From ontology verbalisation to language learning exercises

I’m aware that to most people ‘playing with’ (investigating) ontologies and isiZulu does not sound particularly useful on the face of it. Yet, there’s the some long-term future music, like eventually being able to generate patient discharge notes in one’s own language, which will do its bit to ameliorate the language barrier in healthcare in South Africa so that patients at least will adhere to the treatment instructions a little better, and therewith receive better quality healthcare. But benefits in the short-term might serve something as well. To that end, I proposed an honours project last year, which has been completed in the meantime, and one of the two interesting outcomes has made it into a publication already [1]. As you may have guessed from the title, it’s about automation for language learning exercises. The results will be presented at the 6^th Workshop on Controlled Natural Language, in Maynooth, Ireland in about 2 weeks time (27-28 August). In the remainder of this post, I highlight the main contributions described in the paper.

First, regarding the post’s title, one might wonder what ontology verbalisation has to do with language learning. Nothing, really, except that we could reuse the algorithms from the controlled natural language (CNL) for ontology verbalisation to generate (computer-assisted) language learning exercises whose answers can be computed and marked automatically. That is, the original design of the CNL for things like pluralising nouns, verb conjugation, and negation that is used for verbalising ontologies in isiZulu in theory [2] and in practice [3], was such that the sentence generator is a detachable module that could be plugged in elsewhere for another task that needs such operations.

Practically, the student who designed and developed the back-end, Nikhil Gilbert, preferred Java over Python, so he converted most parts into Java, and added a bit more, notably the ‘singulariser’, a sentence scrabble, and a sentence generator. Regarding the sentence generator, this is used as part of the exercises & answers generator. For instance, we know that humans and the roles they play (father, aunt, doctor, etc.) are mostly in isiZulu’s noun classes 1, 2, 1a, 2a, or 3a, that those classes do not (or rarely?) have non-human nouns and generally it holds for all humans and their roles that they can ‘eat’, ‘talk’ etc. This makes it relatively easy create a noun chain and a verb chain list to mix and match nouns with verbs accordingly (hurrah! for the semantics-based noun class system). Then, with the 231 nouns and 59 verbs in the newly constructed mini-corpus, the noun chain and the verb chain, 39501 unique question sentences could be generated, using the following overall architecture of the system:

Architecture of the CNL-driven CALL system. The arrows indicate which upper layer components make use of the lower layer components. (Source: [1])

From a CNL perspective as well as the language learning perspective, the actual templates for the exercises may be of interest. For instance, when a learner is learning about pluralising nouns and their associated verb, the system uses the following two templates for the questions and answers:

Q: <prefixSG+stem> <SGSC+VerbRoot+FV>
A: <prefixPL+stem> <PLSC+VerbRoot+FV>
Q: <prefixSG+stem> <SGSC+VerbRoot+FV> <prefixSG+stem>
A: <prefixPL+stem> <PLSC+VerbRoot+FV> <prefixPL+stem>

The answers can be generated automatically with the algorithms that generate the plural noun (from ‘prefixSG’ to ‘prefixPL’) and add the plural subject concord (from ‘SGSC’ to ‘PLSC’, in agreement with ‘prefixPL’), which were developed as part of the GeNI project on ontology verbalization. This can then be checked against what the learner has typed. For instance, a generated question could be umfowethu usula inkomishi and the correct answer generated (to check the learner’s response against) is abafowethu basula izinkomishi. Another example is generation of the negation from the positive, or, vv.; e.g.:

Q: <PLSC+VerbRoot+FV>
A: <PLNEGSC+VerbRoot+NEGFV>

For instance, the question may present batotoba and the correct answer is then abatotobi. In total, there are six different types of sentences, with two double, like the plural above, hence a total of 16 templates. It is not a lot, but it turned out it is one of the very few attempts to use a CNL in such way: there is one paper that also will be presented at CNL’18 in the same session [4], and an earlier one [5] uses a fancy grammar system (that we don’t have yet computationally for isiZulu). This is not to be misunderstood as that this is one of the first CNL/NLG-based system for computer-assisted language learning—e.g., there’s assistance in essay writing, grammar concept question generation, reading understanding question generation—but curiously very little on CNLs or NLG for the standard entry-level type of questions to learn the grammar. Perhaps the latter is considered ‘boring’ for English by now, given all the resources. However, thousands of students take introduction courses in isiZulu each year, and some automation can alleviate the pressure of routine activities from the lecturers. We have done some evaluations with learners—with encouraging results—and plan to do some more, so that it may eventually transition to actual use in the courses; that is: TBC…

 

References

[1] Gilbert, N., Keet, C.M. Automating question generation and marking of language learning exercises for isiZulu. 6th International Workshop on Controlled Natural language (CNL’18). IOS Press. Co. Kildare, Ireland, 27-28 August 2018. (in print)

[2] Keet, C.M., Khumalo, L. Toward a knowledge-to-text controlled natural language of isiZulu. Language Resources and Evaluation, 2017, 51(1): 131-157.

[3] Keet, C.M. Xakaza, M., Khumalo, L. Verbalising OWL ontologies in isiZulu with Python. The Semantic Web: ESWC 2017 Satellite Events, Blomqvist, E. et al. (eds.). Springer LNCS vol. 10577, 59-64.

[4] Lange, H., Ljunglof, P. Putting control into language learning. 6th International Workshop on Controlled Natural language (CNL’18). IOS Press. Co. Kildare, Ireland, 27-28 August 2018. (in print)

[5] Gardent, C., Perez-Beltrachini, L. Using FB-LTAG Derivation Trees to Generate Transformation-Based Grammar Exercises. Proc. of TAG+11, Sep 2012, Paris, France. pp117-125, 2012.

Advertisement

ICTs for South Africa’s indigenous languages should be a national imperative, too

South Africa has 11 official languages with English as the language of business, as decided during the post-Apartheid negotiations. In practice, that decision has resulted in the other 10 being sidelined, which holds even more so for the nine indigenous languages, as they were already underresourced. This trend runs counter to the citizens’ constitutional rights and the state’s obligations, as she “must take practical and positive measures to elevate the status and advance the use of these languages” (Section 6 (2)). But the obligations go beyond just language promotion. Take, e.g., the right to have access to the public health system: one study showed that only 6% of patient-doctor consultations was held in the patient’s home language[1], with the other 94% essentially not receiving the quality care they deserve due to language barriers[2].

Learning 3-4 languages up to practical multilingualism is obviously a step toward achieving effective communication, which therewith reduces divisions in society, which in turn fosters cohesion-building and inclusion, and may contribute to achieve redress of the injustices of the past. This route does tick multiple boxes of the aims presented in the National Development Plan 2030. How to achieve all that is another matter. Moreover, just learning a language is not enough if there’s no infrastructure to support it. For instance, what’s the point of searching the Web in, say, isiXhosa when there are only a few online documents in isiXhosa and the search engine algorithms can’t process the words properly anyway, hence, not returning the results you’re looking for? Where are the spellcheckers to assist writing emails, school essays, or news articles? Can’t the language barrier in healthcare be bridged by on-the-fly machine translation for any pair of languages, rather than using the Mobile Translate MD system that is based on canned text (i.e., a small set of manually translated sentences)?

 

Rule-based approaches to develop tools

Research is being carried out to devise Human Language Technologies (HLTs) to answer such questions and contribute to realizing those aspects of the NDP. This is not simply a case of copying-and-pasting tools for the more widely-spoken languages. For instance, even just automatically generating the plural noun in isiZulu from a noun in the singular required a new approach that combined syntax (how it is written) with semantics (the meaning) through inclusion of the noun class system in the algorithms[3] [summary]. In contrast, for English, just syntax-based rules can do the job[4] (more precisely: regular expressions in a Perl script). Rule-based approaches are also preferred for morphological analysers for the regional languages[5], which split each word into its constituent parts, and for natural language generation (NLG). An NLG system generates natural language text from structured data, information, or knowledge, such as data in spreadsheets. A simple way of realizing that is to use templates where the software slots in the values given by the data. This is not possible for isiZulu, because the sentence constituents are context-dependent, of which the idea is illustrated in Figure 1[6].

Figure 1. Illustration of a template for the ‘all-some’ axiom type of a logical theory (structured knowledge) and some values that are slotted in, such as Professors, resp. oSolwazi, and eat, resp. adla and zidla; ‘nc’ denotes the noun class of the noun, which governs agreement across related words in a sentence. The four sample sentences in English and isiZulu represent the same information.

Therefore, a grammar engine is needed to generate even the most basic sentences correctly. The core aspects of the workflow in the grammar engine [summary] are presented schematically in Figure 2[7], which is being extended with more precise details of the verbs as a context-free grammar [summary][8]. Such NLG could contribute to, e.g., automatically generating patient discharge notes in one’s own language, text-based weather forecasts, or online language learning exercises.

Figure 2. The isiZulu grammar engine for knowledge-to-text consists conceptually of three components: the verbalisation patterns with their algorithms to generate natural language for a selection of axiom types, a way of representing the knowledge in a structured manner, and the linking of the two to realize the generation of the sentences on-the-fly. It has been implemented in Python and Owlready.

 

Data-driven approaches that use lots of text

The rules-based approach is known to be resource-intensive. Therefore, and in combination with the recent Big Data hype, data-driven approaches with lost of text are on the rise: it offers the hope to achieve more with less effort, not even having to learn the language, and easier bootstrapping of tools for related languages. This can work, provided one has a lot of good quality text (a corpus). Corpora are being developed, such as the isiZulu National Corpus[9], and the recently established South African Centre for Digital Language Resources (SADiLaR) aims to pool the resources. We investigated the effects of a corpus on the quality of an isiZulu spellchecker [summary], which showed that learning the statistics-driven language model on old texts like the bible does not transfer well to modern-day texts such as news items, nor vice versa[10]. The spellchecker has about 90% accuracy in single-word error detection and it seems to contribute to the intellectualisation[11] of isiZulu [summary][12]. Its algorithms use trigrams and probabilities of their occurrence in the corpus to compute the probability that a word is spelled correctly, illustrated in Figure 3, rather than a dictionary-based approach that is impractical for agglutinating languages. The algorithms were reused for isiXhosa simply by feeding it a small isiXhosa corpus: it achieved about 80% accuracy already even without optimisations.

Figure 3. Illustration of the underlying approach of the isiZulu spellchecker

Data-driven approaches are also pursued in information retrieval to, e.g., develop search engines for isiZulu and isiXhosa[13]. Algorithms for data-driven machine translation (MT), on the other hand, can easily be misled by out-of-domain training data of parallel sentences in both languages from which it has to learn the patterns, such as such as concordial agreement like izi- zi- (see Figure 1). In one of our experiments where the MT system learned from software localization texts, an isiXhosa sentence in the context of health care, Le nto ayiqhelekanga kodwa ngokwenene iyenzeka ‘This is not very common, but certainly happens.’ came out as ‘The file is not valid but cannot be deleted.’, which is just wrong. We are currently creating a domain-specific parallel corpus to improve the MT quality that, it is hoped, will eventually replace the afore-mentioned Mobile Translate MD system. It remains to be seen whether such a data-driven MT or an NLG approach, or a combination thereof, may eventually further alleviate the language barriers in healthcare.

 

Because of the ubiquity of ICTs in all of society in South Africa, HLTs for the indigenous languages have become a necessity, be it for human-human or human-computer interaction. Profit-driven multinationals such as Google, Facebook, and Microsoft put resources into development of HLTs for African languages already. Languages, and the identities and cultures intertwined with them, are a national resource, however; hence, suggesting the need for more research and the creation of a substantial public good of a wide range of HLTs to assist people in the use of their language in the digital age and to contribute to effective communication in society.

[1] Levin, M.E. Language as a barrier to care for Xhosa-speaking patients at a South African paediatric teaching hospital. S Afr Med J. 2006 Oct; 96 (10): 1076-9.

[2] Hussey, N. The Language Barrier: The overlooked challenge to equitable health care. SAHR, 2012/13, 189-195.

[3] Byamugisha, J., Keet, C.M., Khumalo, L. Pluralising Nouns in isiZulu and Related Languages. 17th International Conference on Intelligent Text Processing and Computational Linguistics (CICLing’16). A. Gelbukh (Ed.). Springer LNCS vol 9623, pp. April 3-9, 2016, Konya, Turkey.

[4] Conway, D.M.: An algorithmic approach to English pluralization. In: Salzenberg, C. (ed.) Proceedings of the Second Annual Perl Conference. O’Reilly (1998), San Jose, USA, 17-20 August, 1998

[5] Pretorius, L. & Bosch, S.E. Enabling computer interaction in the indigenous languages of South Africa: The central role of computational morphology. ACM Interactions, 56 (March + April 2003).

[6] Keet, C.M., Khumalo, L. Toward a knowledge-to-text controlled natural language of isiZulu. Language Resources and Evaluation, 2017, 51(1): 131-157.

[7] Keet, C.M. Xakaza, M., Khumalo, L. Verbalising OWL ontologies in isiZulu with Python. The Semantic Web: ESWC 2017 Satellite Events, Blomqvist, E et al. (eds.). Springer LNCS vol 10577, 59-64.

[8] Keet, C.M., Khumalo, L. Grammar rules for the isiZulu complex verb. Southern African Linguistics and Applied Language Studies, 2017, 35(2): 183-200.

[9] L. Khumalo. Advances in Developing corpora in African languages. Kuwala, 2015, 1(2): 21-30.

[10] Ndaba, B., Suleman, H., Keet, C.M., Khumalo, L. The effects of a corpus on isiZulu spellcheckers based on N-grams. In IST-Africa.2016. (May 11-13, 2016). IIMC, Durban, South Africa, 2016, 1-10.

[11] Finlayson, R, Madiba, M. The intellectualization of the indigenous languages of South Africa: Challenges and prospects. Current Issues in Language Planning, 2002, 3(1): 40-61.

[12] Keet, C.M., Khumalo, L. Evaluation of the effects of a spellchecker on the intellectualization of isiZulu. Alternation, 2017, 24(2): 75-97.

[13] Malumba, N., Moukangwe, K., Suleman, H. AfriWeb: A Web Search Engine for a Marginalized Language. Proceedings of 2015 Asian Digital Library Conference, Seoul, South Korea, 9-12 December 2015.

Updated isiZulu spellchecker and new isiXhosa spellchecker

Noting that February is the month of language activism in South Africa and that 21 February is the International Mother Language Day (a United Nations event since 2000), let me add my proverbial two cents to that. Since the launch of the isiZulu spellchecker in November 2016, research and development has progressed quite a bit, so that we have released a new ‘version 2’ of the spellchecker. For those not in-the-know: isiZulu and isiXhosa are both among the 11 official languages of South Africa, with isiZulu the largest language in the country by first language speakers and isiXhosa is slated to make an international breakthrough, as it’s used in the Black Panther movie that was released this weekend. Anyhow, the main novelties of the updated spellchecker are:

• first error correction algorithms for isiZulu;
• improved error detection with a few basic rules, also for isiZulu;
• new isiXhosa error detection and correction;

The source code is open source, and, due to various tool limitations beyond our control, it’s still a standalone jar file (zipped for download). Here’s a screenshot of the tool, where it checks a piece of text from a novel in isiZulu, illustrating that *khupels has a substitution error (khupela was the intended word):

Single word error *khupels that has a substitution error s for a in the intended word (khupela)

The error corrector can propose possible corrections for single-word errors that are either transpositions, substitutions, insertions, or deletions. So, for instance, *eybo, *yrbo, *yeebo, and *ybo, respectively, cf. the correctly spelled yebo ‘yes’. It doesn’t perform equally well on each type of typo yet, with the best results obtained for transpositions. As with the error detector, it relies on a data-driven approach, with, for error correction, a lot more statistics-based algorithms cf. the error detection-only algorithms. They are described in detail in Frida Mjaria’s 2017 CS honours project. Suggestion accuracy (i.e., that it at least can suggest something) is 95% and suggestion relevance (that it contains the intended word) made it to 61%, mainly due to weak results of corrections for insertion errors (they mess too much with the trigrams).

The error detection accuracy has been improved mainly through better handling of punctuation, thanks to Norman Pilusa’s programming efforts. This was done through a series of rules on top of the data-driven approach, for it is too hard to learn those from a large corpus. For instance, semi-colons, end-of-sentence periods, and numbers (written in isiZulu like, e.g., ngu-42 rather than just 42) are now mostly ignored rather than the words adjacent to it being detected as probably misspelt. It works better than spellchecker.net’s version, which is the only other available isiZulu spellchecker: on a random selection of actual pieces of text, our tool obtained 91.71% lexical recall for error detection, whereas the spellchecker.net’s version got to 82.66% on the same text. Put differently: spellchecker.net flagged about twice as many words as incorrect as ours did (so there wasn’t much point in comparing error corrections).

Finally, because all the algorithms are essentially language-independent (ok, there’s an underlying assumption of using them for highly agglutinative languages), we fed the algorithms a large isiXhosa corpus that is being developed as part of another project, and incorporated that into the spellchecker. There’s room for some fine-tuning especially for the corrector, but at least now there is one, thanks to Norman Pilusa’s software development contributions. That we thought we could get away with this approach is thanks to Nthabiseng Mashiane’s 2017 CS honours project, which showed that the results would be fairly good (>80% error detection) with more data. We also tried a rules-based approach for isiXhosa. It obtained better accuracies than the statistical language model of Nthabiseng, but only for those parts of speech covered by the rules, which is a subset of all types of words. If you’re interested in those rules, please check out Siseko Neti’s 2017 CS Honours project. To the best of my knowledge, it’s the first time those rules have been formally represented in a computer-usable format and they may be useful for other endeavours, such as morphological analysers.

A section of the isiXhosa Wikipedia entry about the UN (*ukuez should be ukuze, which is among the proposed words).

Further improvements are possible, which are being scoped for a v3 some time later. For instance, for the linguists and language scholars: what are the most common typos? What are the most commonly used words? If we had known that, it would have been an easy way to boost the performance. Can we find optimisations to substitutions, insertions, and deletions similar to the one for transpositions? Should some syntax rules be added for further optimisation? These are some of the outstanding questions. If you’re interested in that or related questions, or you would like to use the algorithms in your tool, please contact me.

The isiZulu spellchecker seems to contribute to ‘intellectualisation’ of isiZulu

Perhaps putting ‘intellectualisation’ in sneer quotes isn’t nice, but I still find it an odd term to refer to a process of (in short, from [1]) coming up with new vocabulary for scientific speech, expression, objective thinking, and logical judgments in a natural language. In the country I grew up, terms in our language were, and still are, invented more because of a push against cultural imperialism and for home language promotion rather than some explicit process to intellectualise the language in the sense of “let’s invent some terms because we need to talk about science in our own language” or “the language needs to grow up” sort of discourses. For instance, having introduced the beautiful word geheugensanering (NL) that captures the concept of ‘garbage collection’ (in computing) way better than the English joke-term for it, elektronische Datenverarbeitung (DE) for ‘ICT’, técnicas de barrido (ES) for ‘sweep line’ algorithms, and mot-dièse (FR) for [twitter] ‘hashtag’, to name but a few inventions.

Be that as it may, here in South Africa, it goes under the banner of intellectualisation, with particular reference to the indigenous languages [2]; e.g., having introduced umakhalekhukhwini ‘cell/mobile phone’ (decomposed: ‘the thing that rings in your pocket’) and ukudlulisa ikheli for ‘pass by reference’ in programming (longer list of isiZulu-English computing and ICT terms), which is occurring for multiple subject domains [3]. Now I ended up as co-author of a paper that has ‘intellectualisation’ in its title [4]: Evaluation of the effects of a spellchecker on the intellectualization of isiZulu that appeared just this week in the Alternation journal.

The main general question we sought to answer was whether human language technologies, and in particular the isiZulu spellchecker launched last year, contribute to the language’s intellectualisation. More specifically, we aimed to answer the following three questions:

1. Is the spellchecker meeting end-user needs and expectations?
2. Is the spellchecker enabling the intellectualisation of the language?
3. Is the lexicon growing upon using the spellchecker?

The answers in a nutshell are: 1) yes, the spellchecker does meet end-user needs and expectations (but there are suggestions further improving its functionality), 2) users perceive that the spellchecker enables the intellectualisation of the language, and 3) non-dictionary words were added, i.e., the lexicon is indeed growing.

The answer to the last question provides some interesting data for linguists to bite their teeth in. For instance, a user had added to the spellchecker’s dictionary LikaSekelaShansela, which is an inflected form of isekelashansela ‘Vice Chancellor’ (that is recognised as correct by the spellchecker). Also some inconsistencies—from a rule-of-thumb viewpoint—in word formation were observed; e.g., usosayensi ‘scientist’ vs. unompilo ‘nurse’. If one were to follow consistently the word formation process for various types of experts in isiZulu, such as usosayensi ‘scientist’, usolwazi ‘professor’, and usomahlaya ‘comedian’, then one reasonably could expect ‘nurse’ to be *usompilo rather than unompilo. Why it isn’t, we don’t know. Regardless, the “add to dictionary” option of the spellchecker proved to be a nice extra feature for a data-driven approach to investigate intellectualisation of a language.

Version 1 of the isiZulu spellchecker that was used in the evaluation was ok and reasonably could not have interfered negatively with any possible intellectualisation (average SUS score of 75 and median 82.5, so ‘good’). It was ok in the sense that a majority of respondents thought that the entire tool was helpful, no features should be removed, it enhances their work, and so on (see paper for details). For the software developers among you who have spare time: they’d like, mainly, to have it as a Chrome and MS Word plugin, predictive text/autocomplete, and have it working on the mobile phone. The spellchecker has improved in the meantime thanks to two honours students, and I will write another blog post about that next.

As a final reflection: it turned out there isn’t a way to measure the level of intellectualisation in a ‘hard sciences’ way, so we concluded the other answers based on data that came from the somewhat fluffy approach of a survey and in-depth interviews (a ‘mixed-methods’ approach, to give it a name). It would be nice to have a way to measure it, though, so one would be able to say which languages are more or less intellectualised, what level of intellectualisation is needed to have a language as language of instruction and science at tertiary level of education and for dissemination of scientific knowledge, and to what extent some policy x, tool y, or activity z contributes to the intellectualization of a language.

 

References

[1] Havránek, B. 1932. The functions of literary language and its cultivation. In Havránek, B and Weingart, M. (Eds.). A Prague School Reader on Esthetics, Literary Structure and Style. Prague: Melantrich: 32-84.

[2] Finlayson, R, Madiba, M. The intellectualization of the indigenous languages of South Africa: Challenges and prospects. Current Issues in Language Planning, 2002, 3(1): 40-61.

[3]Khumalo, L. Intellectualization through terminology development. Lexikos, 2017, 27: 252-264.

[4] Keet, C.M., Khumalo, L. Evaluation of the effects of a spellchecker on the intellectualization of isiZulu. Alternation, 2017, 24(2): 75-97.

A grammar of the isiZulu verb (present tense)

If you have read any of the blog posts on (automated) natural language generation for isiZulu, then you’ll probably agree with me that isiZulu verbs are non-trivial. True, verbs in other languages are most likely not as easy as in English, or Afrikaans for that matter (e.g., they made irregular verbs regular), but there are many little ‘bits and pieces’ ‘glued’ onto the verb root that make it semantically a ‘heavy’ element in a sentence. For instance:

• Aba-shana ba-ya-zi-theng-is-el-an-a                izimpahla
• Children   2.SC-Pres-8.OC-buy_VR -C-A-R-FV 8.clothes
• ‘The children are selling the clothes to each other’

The ba is the subject concord (~conjugation) to match with the noun class (which is 2) of the noun that plays the subject in the sentence (abashana), the ya denotes a continuous action (‘are doing something’ in the present), the zi is the object concord for the noun class (8) of the noun that plays the object in the sentence (izimpahla), theng is the verb root, then comes the CARP extension with is the causative (turning ‘buy’ into ‘sell’), and el the applicative and an the reciprocative, which take care of the ‘to each other’, and then finally the final vowel a.

More precisely, the general basic structure of the verb is as follows:

where NEG is the negative; SC the subject concord; T/A denotes tense/aspect; MOD the mood; OC the object concord; Verb Rad the verb radical; C the causative; A the applicative; R the reciprocal; and P the passive. For instance, if the children were not selling the clothes to each other, then instead of the SC, there would be the NEG SC in that position, making the verb abayazithengiselana.

To make sense of all this in a way that it would be amenable to computation, we—my co-author Langa Khumalo and I—specified the grammar of the complex verb for the present tense in a CFG using an incremental process of development. To the best of our (and the reviewer’s) knowledge, the outcome of the lengthy exercise is (1) the first comprehensive and precisely formulated documentation of the grammar rules for the isiZulu verb present tense, (2) all together in one place (cf. fragments sprinkled around in different papers, Wikipedia, and outdated literature (Doke in 1927 and 1935)), and (3) goes well beyond handling just one of the CARP, among others. The figure below summarises those rules, which are explained in detail in the forthcoming paper “Grammar rules for the isiZulu complex verb”, which will be published in the Southern African Linguistics and Applied Language Studies [1] (finally in print, yay!).

It is one thing to write these rules down on paper, and another to verify whether they’re actually doing what they’re supposed to be doing. Instead of fallible and laborious manual checking, we put them in JFLAP (for the lack of a better alternative at the time; discussed in the paper) and tested the CFG both on generation and recognition. The tests went reasonably well, and it helped fixing a rule during the testing phase.

Because the CFG doesn’t take into account phonological conditioning for the vowels, it generates strings not in the language. Such phonological conditioning is considered to be a post-processing step and was beyond the scope of elucidating and specifying the rules themselves. There are other causes of overgeneration that we did not get around to doing, for various reasons: there are rules that go across the verb root, which are simple to represent in coding-style notation (see paper) but not so much in a CFG, and rules for different types of verbs, but there’s no available resource that lists which verb roots are intransitive, which as monosyllabic and so on. We have started with scoping rules and solving issues for the latter, and do have a subset of phonological conditioning rules; so, to be continued… For now, though, we have completed at least one of the milestones.

Last, but not least, in case you wonder what’s the use of all this besides the linguistics to satisfy one’s curiosity and investigate and document an underresourced language: natural language generation for intelligent user interfaces in localised software, spellcheckers, and grammar checkers, among others.

 

References

[1] Keet, C.M., Khumalo, L. Grammar rules for the isiZulu complex verb. Southern African Linguistics and Applied Language Studies, (in print). Submitted version (the rules are the same as in the final version)

Aligning different relations: the case of part-whole relations—LDK2017

Despite the best intentions, I did not get around to writing a post on the paper that I presented last week at the First International Conference on Language, Data and Knowledge 2017, 19-20 June, Galway, Ireland, and now Paul Groth also ‘beat’ me to it writing a nice conference report of it. On the bright side, it is an opportunity to say upfront I really enjoyed the conference and look forward to the next edition in 2019. The ESWC’17 organisers might be slightly disappointed that there was no special track on the multilingual semantic web after all, but I did get the distinct impression that the LDK17 authors might just all have gambled on LDK17—an opportunity to binge two days on all things natural language & Semantic Web—rather than on one track at an overpriced conference (despite the allure of it being A-rated).

So, what was my paper about that could have been submitted to either? I ended up struggling—and solving—an issue with aligning OWL object properties that were not simple 1:1 mappings, in a similar scope as our ESWC17 paper (introduced here) [4], but then with too many complications. Complications were due to the different conceptualisations of part-whole relations and that one of the requirements was to solve what to do with an object property (relation, relationship) that does not have a neat, single, label, and therewith neither fitting with the common OWL modelling paradigm nor with the recently agreed-upon ontolex-lemon model for linguistic annotations.

The start of all this sounded nice and doable: we need to generate natural language for healthcare, using, e.g., SNOMED CT, in local languages in South Africa, focussing on the largest one, being isiZulu. Medical terminologies are riddled with part-whole relations, so we sought to address that one (simple existentials already having been solved), availing of a standard list of part-whole relations (e.g. [1]). That turned out to be a non-trivial exercise, but doable eventually [2]. What wasn’t addressed in [2] was that some ‘common’ part-whole relations, such as membership and containment, weren’t like that in isiZulu, at all. Moreover, it wasn’t just a language issue, but ontological as well. The LDK17 paper “Representing and aligning similar relations: parts and wholes in isiZulu vs English” [3] describes this in some detail.

Here’s a (simplified) list of (assumed to be) common part-whole relations, which takes into account both transitivity differences and domain and range:

Now here’s the one based on the isiZulu language and some ontological analysis of that:

That is: there are both generalisations—some distinctions are not being made—and specialisations—some distinctions are made here but not elsewhere. For instance, ‘musician is part of some orchestra’ and ‘heart is part of some human’ (or vv.) is all done and described in the same way (ingxenye ‘part of’ and SC+CONJ for ‘has part’ [more about that below]). Yet, there is a difference between an individual (e.g., a voter) participating in some process and a collective (e.g., the electorate) participating in a process, or vv. The paper describes this more precisely, going into some detail regarding the differences in categories of domain and range and into the consequences on transitivity of mereological parthood.

The other ‘odd thing’—cf. current multilingual Semantic Web assumptions and technologies, that is—is that while the conceptualisation of ‘has part’ exists, it does not have a single label as in English (or in several other languages, such as heeft as deel), but it is dependent on the noun class of the noun of the class that play the part and play the whole in the relation. It combines the subject concord (~conjugation) of the noun class of the noun that plays the whole with a conjunction that is phonologically conditioned based on the first letter of the noun that plays the part; with verbalisation in the plural and three phonological cases, there are 18 possible strings all denoting ‘has part’. This still could be sorted with a language with inverses, provided the part-of direction has a name, like the ingxenye. This is not the case for containment, however. Instead of the relation (object property) having a name—be this a verb like ‘contained in’ or some noun phrase—it is the noun that plays the whole (the container, if you will) that gets modified. For instance, imvilophu ‘envelope’ and emvilophini denoting ‘contained in the envelope’, or, for individuals and locations, the city iTheku ‘Durban’ and eThekwini meaning ‘located in Durban’ (no typo—there’s some phonological conditioning I’m brushing over). While I have gotten used to such constructions, it generated some surprise among several attendees that one can have notions, concepts, views on or interpretations or descriptions of reality, that exist but do not have even one single string of text throughout to refer to regardless the context it is used.

The naming issue was solved by adding some arbitrary string as ‘name’ of the object property, and relating that to the function that verbalises that specific part-whole relation. The former issue, i.e., not all the same part-whole relations, required a bit more work, using ontology pattern alignments, by extending one correspondence pattern from the ODP catalogue and introducing a new one (see paper for the formal details), using the same broad framework of formalisation as proposed in [4].

All this was then implemented and aligned, and verified to not result in some unsatisfiable classes, object properties, or inconsistency (files). This also works in the isiZulu verbalisation tool we demo-ed at ESWC17 (described in the previous post) [5], all as part of the NRF-funded GeNI project.

Now, ideally, I already would have had the time to read the papers I flagged in my LDK17 conference notes with “check paper”. I haven’t yet due to end-of-semester tasks. So, on the basis of just a positive-seeming presentation, here are a few that are on the top of my list to check out first, for quite different reasons:

• Interaction between natural language reading capabilities and math education, focusing on language production (i.e., ‘can you talk about it?’) [6], mainly because math education in South Africa faces a lot of problems. It also generated a lively discussion in the Q&A session.
• The OnLiT ontology for linguistic [7] and LLODifying linguistic glosses [8] terminology (also: one of the two also won the best paper award).
• Deep text generation, for it was looking at trying to address skewed or limited data to learn from [9], which is an issue we face when trying to do some NLP with most South African languages.

 

References

[1] Keet, C.M., Artale, A. Representing and Reasoning over a Taxonomy of Part-Whole Relations. Applied Ontology, 2008, 3(1-2):91-110.

[2] Keet, C.M., Khumalo, L. On the verbalization patterns of part-whole relations in isiZulu. 9th International Natural Language Generation conference (INLG’16), September 5-8, 2016, Edinburgh, UK. ACL.

[3] Keet, C.M. Representing and aligning similar relations: parts and wholes in isiZulu vs English. In: Gracia J., Bond F., McCrae J., Buitelaar P., Chiarcos C., Hellmann S. (eds) Language, Data, and Knowledge LDK 2017. Springer LNAI vol 10318, 58-73.

[4] Fillottrani, P.R., Keet, C.M. Patterns for Heterogeneous TBox Mappings to Bridge Different Modelling Decisions. 14th Extended Semantic Web Conference (ESWC’17). Springer LNCS. Portoroz, Slovenia, May 28 – June 2, 2017.

[5] Keet, C.M. Xakaza, M., Khumalo, L. Verbalising OWL ontologies in isiZulu with Python. 14th Extended Semantic Web Conference (ESWC’17). Springer LNCS. Portoroz, Slovenia, May 28 – June 2, 2017. (demo paper)

[6] Crossley, S., Kostyuk, V. Letting the genie out of the lamp: using natural language processing tools to predict math performance. In: Gracia J., Bond F., McCrae J., Buitelaar P., Chiarcos C., Hellmann S. (eds) Language, Data, and Knowledge LDK 2017. Springer LNAI vol 10318, 330-342.

[7] Klimek, B., McCrae, J.P., Lehmann, C., Chiarcos, C., Hellmann, S. OnLiT: and ontology for linguistic terminology. In: Gracia J., Bond F., McCrae J., Buitelaar P., Chiarcos C., Hellmann S. (eds) Language, Data, and Knowledge LDK 2017. Springer LNAI vol 10318, 42-57.

[8] Chiarcos, C., Ionov, M. Rind-Pawlowski, M., Fäth, C., Wichers Schreur, J., Nevskaya. I. LLODifying linguistic glosses. In: Gracia J., Bond F., McCrae J., Buitelaar P., Chiarcos C., Hellmann S. (eds) Language, Data, and Knowledge LDK 2017. Springer LNAI vol 10318, 89-103.

[9] Dethlefs N., Turner A. Deep Text Generation — Using Hierarchical Decomposition to Mitigate the Effect of Rare Data Points. In: Gracia J., Bond F., McCrae J., Buitelaar P., Chiarcos C., Hellmann S. (eds) Language, Data, and Knowledge LDK 2017. Springer LNAI vol 10318, 290-298.

Our ESWC17 demos: TDDonto2 and an OWL verbaliser for isiZulu

Besides the full paper on heterogeneous alignments for 14th Extended Semantic Web Conference (ESWC’17) that will take place next week in Portoroz, Slovenia, we also managed to squeeze out two demo papers. You may already know of TDDonto2 with Kieren Davies and Agnieszka Lawrynowicz, which was discussed in an earlier post that has been updated with a tutorial video. It now has a demo paper as well [1], which describes the rationale and a few scenarios. The other demo, with Musa Xakaza and Langa Khumalo, is new-new, but the regular reader might have seen it coming: we finally managed to link the verbalisation patterns for certain Description Logic axiom types [2,3] to those in OWL ontologies. The tool takes as input an ontology in isiZulu and the verbalisation algorithms, and out come the isiZulu sentences, be this in plain text for further processing or in a GUI for inspection by a domain expert [4]. There is a basic demo-screencast to show it’s all working.

The overall architecture may be of interest, for it deviates from most OWL verbalisers. It is shown in the following figure:

For instance, we use the Python-based OWL API Owlready, rather than a Java-based app, for Python is rather popular in NLP and the verbalisation algorithms may be used elsewhere as well. We made more such decisions with the aim to make whatever we did as multi-purpose usable as possible, like the list of nouns with noun classes (surprisingly, and annoyingly, there is no such readily available list yet, though isizulu.net probably will have it somewhere but inaccessible), verb roots, and exceptions in pluralisation. (Problems for integrating the verbaliser with, say, Protégé will be interesting to discuss during the demo session!)

The text-based output doesn’t look as nice as the GUI interface, so I will show here only the GUI interface, which is adorned with some annotations to illustrate that those verbalisation algorithms in the background are far from trivial templates:

For instance, while in English the universal quantification is always ‘Each’ or ‘All’ regardless the named class quantified over, in isiZulu it depends on the noun class of the noun that is the name of the OWL class. For instance, in the figure above, izingwe ‘leopards’ is in noun class 10, so the ‘Each/All’ is Zonke, amavazi ‘vases’ is in noun class 6, so ‘Each/All’ then becomes Onke, and abantu ‘people’/’humans’ is in noun class 2, making Bonke. There are 17 noun classes. They also determine the subject concords (SC, alike conjugation) for the verbs, with zi- for noun class 10, ­a- for noun class 6, and ba- for noun class 2, to name a few. How this all works is described in [2,3]. We’ve implemented all those algorithms and integrated the pluraliser [5] in it to make it work. The source files are available to check and play with already, you can do so and ask us during the ESWC17 demo session, and/or also have a look at the related outputs of the NRF-funded project Grammar Engine for Nguni natural language interfaces (GeNi).

 

References

[1] Davies, K. Keet, C.M., Lawrynowicz, A. TDDonto2: A Test-Driven Development Plugin for arbitrary TBox and ABox axioms. Extended Semantic Web Conference (ESWC’17), Springer LNCS. Portoroz, Slovenia, May 28 – June 2, 2017. (demo paper)

[2] Keet, C.M., Khumalo, L. Toward a knowledge-to-text controlled natural language of isiZulu. Language Resources and Evaluation, 2017, 51:131-157.

[3] Keet, C.M., Khumalo, L. On the verbalization patterns of part-whole relations in isiZulu. 9th International Natural Language Generation conference (INLG’16), 5-8 September, 2016, Edinburgh, UK. Association for Computational Linguistics, 174-183.

[4] Keet, C.M. Xakaza, M., Khumalo, L. Verbalising OWL ontologies in isiZulu with Python. 14th Extended Semantic Web Conference (ESWC’17). Springer LNCS. Portoroz, Slovenia, May 28 – June 2, 2017. (demo paper)

[5] Byamugisha, J., Keet, C.M., Khumalo, L. Pluralising Nouns in isiZulu and Related Languages. 17th International Conference on Intelligent Text Processing and Computational Linguistics (CICLing’16), Springer LNCS. April 3-9, 2016, Konya, Turkey.

Launch of the isiZulu spellchecker

Langa Khumalo, ULPDO director, giving the spellchecker demo, pointing out a detected spelling error in the text. On his left, Mpho Monareng, CEO of PanSALB.

Yesterday, the isiZulu spellchecker was launched at UKZN’s “Launch of the UKZN isiZulu Books and Human Language Technologies” event, which was also featured on 702 live radio, SABC 2 Morning Live, and e-news during the day. What we at UCT have to do with it is that both the theory and the spellchecker tool were developed in-house by members of the Department of Computer Science at UCT. The connection with UKZN’s University Language Planning & Development Office is that we used a section of their isiZulu National Corpus (INC) [1] to train the spellchecker with, and that they wanted a spellchecker (the latter came first).

The theory behind the spellchecker was described briefly in an earlier post and it has been presented at IST-Africa 2016 [2]. Basically, we don’t use a wordlist + rules-based approach as some experiments of 20 years ago did, nor a wordlist + a few rules of the now-defunct translate.org.za OpenOffice v3 plugin seven years ago, but a data-driven approach with a statistical language model that uses tri-grams. The section of the INC we used were novels and news items, so, including present-day isiZulu texts. At the time of the IST-Africa’16 paper, based on Balone Ndaba’s BSc CS honours project, the spell checking was very proof-of-concept, but it showed that it could be done and still achieve a good enough accuracy. We used that approach to create an enduser-usable isiZulu spellchecker, which saw the light of day thanks to our 3rd-year CS@UCT student Norman Pilusa, who both developed the front-end and optimised the backend so that it has an excellent performance.

Upon starting the platform-independent isiZulu_spellchecker.jar file, the English interface version looks like this:

You can write text in the text box, or open a txt or docx file, which then is displayed in the textbox. Click “Run”. Now there are two options: you can choose to step-through the words that are detected as misspelled one at a time or “Show All” words that are detected as misspelled. Both are shown for some sample text in the screenshot below.

processing one error at a time

highlighting all words detected as very probably misspelled

Then it is up to you to choose what to do with it: correct it in the textbox, “Ignore once”, “Ignore all”, or “Add” the word to your (local) dictionary. If you have modified the text, you can save it with the changes made by clicking “Save correction”. You also can switch the interface from the default English to isiZulu by clicking “File – Use English”, and back to English via “iFayela – ulimi lesingisi”. You can download the isiZulu spellchecker from the ULPDO website and from the GitHub repository for those who want to get their hands on the source code.

To anticipate some possible questions you may have: incorporating it as a plugin to Microsoft word, OpenOffice/LibreOffice, and Mozilla Firefox was in the planning. The former is technologically ‘closed source’, however, and the latter two have a certain way of doing spellchecking that is not amenable to the data-driven approach with the trigrams. So, for now, it is a standalone tool. By design, it is desktop-based rather than for mobile phones, because according to the client (ULPDO@UKZN), they expect the first users to be professionals with admin documents and emails, journalists writing articles, and such, writing on PCs and laptops.

There was also a trade-off between a particular sort of error: the tool now flags more words as probably incorrect than it could have, yet it will detect (a subset of) capitalization, correctly, such as KwaZulu-Natal whilst flagging some of the deviant spellings that go around, as shown in the screenshot below.

The customer preferred recognising such capitalisation.

Error correction sounds like an obvious feature as well, but that will require a bit more work, not just technologically, but also the underlying theory. It will probably be an honours project topic for next year.

In the grand scheme of things, the current v1 of the spellchecker is only a small step—yet, many such small steps in succession will get one far eventually.

The launch itself saw an impressive line-up of speeches and introductions: the keynote address was given by Dr Zweli Mkhize, UKZN Chancellor and member of the ANC NEC; Prof Ramesh Krishnamurthy, from Aston University UK, gave the opening address; Mpho Monareng, CEO of PanSALB gave an address and co-launched the human language technologies; UKZN’s VC Andre van Jaarsveld provided the official welcome; and two of UKZN’s DVCs, Prof Renuka Vithal and Prof Cheryl Potgieter, gave presentations. Besides our ‘5-minutes of fame’ with the isiZulu spellchecker, the event also launched the isiZulu National Corpus, the isiZulu Term Bank, the ZuluLex mobile-compatible application (Android and iPhone), and two isiZulu books on collected short stories and an English-isiZulu architecture glossary.

 

References

[1] Khumalo, L. Advances in developing corpora in African languages. Kuwala, 2015, 1(2): 21-30.

[2] Ndaba, B., Suleman, H., Keet, C.M., Khumalo, L. The Effects of a Corpus on isiZulu Spellcheckers based on N-grams. IST-Africa 2016. May 11-13, 2016, Durban, South Africa.

Relations with roles / verbalising object properties in isiZulu

The narratives can be very different for the paper “A model for verbalising relations with roles in multiple languages” that was recently accepted paper at the 20th International Conference on Knowledge Engineering and Knowledge management (EKAW’16), for the paper makes a nice smoothie of the three ingredients of language, logic, and ontology. The natural language part zooms in on isiZulu as use case (possibly losing some ontologist or logician readers), then there are the logics about mapping the Description Logic DLR’s role components with OWL (lose possible interest of the natural language researchers), and a bit of philosophy (and lose most people…). It solves some thorny issues when trying to verbalise complicated verbs that we need for knowledge-to-text natural language generation in isiZulu and some other languages (e.g., German). And it solves the matching of logic-based representations popularised in mainly UML and ORM (that typically uses a logic in the DLR family of Description Logic languages) with the more commonly used OWL. The latter is even implemented as a Protégé plugin.

Let me start with some use-cases that cause problems that need to be solved. It is well-known that natural language renderings of ontologies facilitate communication with domain experts who are expected to model and validate the represented knowledge. This is doable for English, with ACE in the lead, but it isn’t for grammatically richer languages. There, there are complications, such as conjugation of verbs, an article that may be dependent on the preposition, or a preposition may modify the noun. For instance, works for, made by, located in, and is part of are quite common names for object properties in ontologies. They all do have a dependent preposition, however, there are different verb tenses, and the latter has a copulative and noun rather than just a verb. All that goes into the object properties name in an ‘English-based ontology’ and does not really have to be processed further in ontology verbalisation other than beautification. Not so in multiple other languages. For instance, the ‘in’ of located in ends up as affixes to the noun representing the object that the other object is located in. Like, imvilophu ‘envelope’ and emvilophini ‘in the envelope’ (locative underlined). Even something straightforward like a property eats can end up having to be conjugated differently depending on who’s eating: when a human eats, it is udla in isiZulu, but for, say, a dog, it is idla (modification underlined), which is driven by the system of noun classes, of which there are 17 in isiZulu. Many more examples illustrating different issues are described in the paper. To make a long story short, there are gradations in complicating effects, from no effect where a preposition can be squeezed in with the verb in naming an OP, to phonological conditioning, to modifying the article of the noun to modifying the noun. A ‘3rd pers. sg.’ may thus be context-dependent, and notions of prepositions may modify the verb or the noun or the article of the noun, or both. For a setting other than English ontologies (e.g., Greek, German, Lithuanian), a preposition may belong neither to the verb nor to the noun, but instead to the role that the object plays in the relation described by the verb in the sentence. For instance, one obtains yomuntu, rather than the basic noun umuntu, if it plays the role of the whole in a part-whole relation like in ‘heart is part of a human’ (inhliziyo iyingxenye yomuntu).

The question then becomes how to handle such a representation that also has to include roles? This is quite common in conceptual data modelling languages and in the DLR family of DL languages, which is known in ontology as positionalism [2]. Bumping up the role to an element in the representation language—thus, in addition to the relationship—enables one to attach information to it, like whether there is a (deep) preposition associated with it, the tense, or the case. Such role-based annotations can then be used to generate the right element, like einen Betrieb ‘some company’ to adjust the article for the case it goes with in German, or ya+umuntu=yomuntu ‘of a human’, modifying the noun in the object position in the sentence.

To get this working properly, with a solid theoretical foundation, we reused a part of the conceptual modelling languages’ metamodel [3] to create a language model for such annotations, in particular regarding the attributes of the classes in the metamodel. On its own, however, it is rather isolated and not immediately useful for ontologies that we set out to be in need of verbalising. To this end, it links to the ‘OWL way of representing relations’ (ontologically: the so-called standard view), and we separate out the logic-based representation from the readings that one can generate with the structured representation of the knowledge. All in all, the simplified high-level model looks like the picture below.

Simplified diagram in UML Class Diagram notation of the main components (see paper for attributes), linking a section of the metamodel (orange; positionalist commitment) to predicates (green; standard view) and their verbalisation (yellow). (Source: [1])

That much for the conceptual part; more details are described in the paper.

Just a fluffy colourful diagram isn’t enough for a solid implementation, however. To this end, we mapped one of the logics that adhere to positionalism to one of the standard view, being DLR [4] and OWL, respectively. It equally well could have been done for other pairs of languages (e.g., with Common Logic), but these two are more popular in terms of theory and tools.

Having the conceptual and logical foundations in place, we did implement it to see whether it actually can be done and to check whether the theory was sufficient. The Protégé plugin is called iMPALA—it could be an abbreviation for ‘Model for Positionalism And Language Annotation’—that both writes all the non-OWL annotations in a separate XML file and takes care of the renderings in Protégé. It works; yay. Specifically, it handles the interaction between the OWL file, the positionalist elements, and the annotations/attributes, plus the additional feature that one can add new linguistic annotation properties, so as to cater for extensibility. Here are a few screenshots:

OWL’s arbeitetFuer ‘works for’ is linked to the relationship arbeiten.

The prey role in the axiom of the impala being eaten by the ibhubesi.

Annotations of the prey role itself, which is a role in the relationship ukudla.

We did test it a bit, from just the regular feature testing to the African Wildlife ontology that was translated into isiZulu (spoken in South Africa) and a people and pets ontology in ciShona (spoken in Zimbabwe). These details are available in the online supplementary material.

The next step is to tie it all together, being the verbalisation patterns for isiZulu [5,6] and the OWL ontologies to generate full sentences, correctly. This is set to happen soon (provided all the protests don’t mess up the planning too much). If you want to know more details that are not, or not clearly, in the paper, then please have a look at the project page of A Grammar engine for Nguni natural language interfaces (GeNi), or come visit EKAW16 that will be held from 21-23 November in Bologna, Italy, where I will present the paper.

 

References

[1] Keet, C.M., Chirema, T. A model for verbalising relations with roles in multiple languages. 20th International Conference on Knowledge Engineering and Knowledge Management EKAW’16). Springer LNAI, 19-23 November 2016, Bologna, Italy. (in print)

[2] Leo, J. Modeling relations. Journal of Philosophical Logic, 2008, 37:353-385.

[3] Keet, C.M., Fillottrani, P.R. An ontology-driven unifying metamodel of UML Class Diagrams, EER, and ORM2. Data & Knowledge Engineering, 2015, 98:30-53.

[4] Calvanese, D., De Giacomo, G. The Description Logics Handbook: Theory, Implementation and Applications, chap. Expressive description logics, pp. 178-218. Cambridge University Press (2003).

[5] Keet, C.M., Khumalo, L. Toward a knowledge-to-text controlled natural language of isiZulu. Language Resources and Evaluation, 2016, in print.

[6] Keet, C.M., Khumalo, L. On the verbalization patterns of part-whole relations in isiZulu. Proceedings of the 9th International Natural Language Generation conference 2016 (INLG’16), Edinburgh, Scotland, Sept 2016. ACL, 174-183.

Surprising similarities and differences in orthography across several African languages

It is well-known that natural language interfaces and tools in one’s own language are known to be useful in ICT-mediated communication. For instance, tools like spellcheckers and Web search engines, machine translation, or even just straight-forward natural language processing to at least ‘understand’ documents and find the right one with a keyword search. Most languages in Southern Africa, and those in the (linguistically called) Bantu language family, are still under-resourced, however, so this is not a trivial task due to the limited data and researched and documented grammar. Any possibility to ‘bootstrap’ theory, techniques, and tools developed for one language and to fiddle just a bit to make it work for a similar one will save many resources compared to starting from scratch time and again. Likewise, it would be very useful if both the generic and the few language-specific NLP tools for the well-resourced languages could be reused or easily adapted across languages. The question is: does that work? We know very little about whether it does. Taking one step back, then: for that bootstrapping to work well, we need to have insight into how similar the languages are. And we may be able to find that out if only we knew how to measure similarity of languages.

The most well-know qualitative way for determining some notion of similarity started with Meinhof’s noun class system [1] and the Guthrie zones. That’s interesting, but not nearly enough for computational tools. An experiment has been done for morphological analysers [2], with promising results, yet it also had more of a qualitative flavour to it.

I’m adding here another proverbial “2 cents” to it, by taking a mostly quantitative approach to it, and focusing on orthography (how things are written down) in text documents and corpora. This was a two-step process. First, 12 versions of the Universal Declaration of Human Rights were examined on tokens and their word length; second, because the UDHR is a quite small document, isiZulu corpora were examined to see whether the UDHR was a representative sample, i.e., whether extrapolation from its results may be justified. The methods, results, and discussion are described in “An assessment of orthographic similarity measures for several African languages” [3].

The really cool thing of the language comparison is that it shows clusters of languages, indicating where bootstrapping may have more or less success, and they do not quite match with Guthrie zones. The cumulative frequency distributions of the words in the UDHR of several languages spoken in Sub-Saharan Africa is shown in the figure below, where the names of the languages are those of the file names of the NLTK data kit that contains the quality translations of the UDHR.

Cumulative frequency distributions of the words in the UDHR of several languages spoken in Sub-Saharan Africa (Source: [3]).

The paper contains some statistical tests, showing that the bottom cluster are not statistically significantly different form each other, but they are from the ‘middle’ cluster. So, the word length distribution of Kiswahili is substantially different from that of, among others, isiZulu, in that it has more shorter words and isiZulu more longer words, but Kiswahili’s pattern is similar to that of Afrikaans and English. This is important for NLP, for isiZulu is known to be highly agglutinating, but English (and thus also Kiswahili) is disjunctive. How important is such a difference? The simple answer is that grammatical elements of a sentences get ‘glued’ together in isiZulu, whereas at least some of them are written as separate words in Kiswahili. This is not to be conflated with, say, German, Dutch, and Afrikaans, where nouns can be concatenated to form new words, but, e.g., a preposition is glued onto a noun. For instance, ‘of clay’ is ngobumba, contracting nga+ubumba with a vowel coalescence rule (-a + u- = -o-), which thus happens much less often in a language with disjunctive orthography. This, in turn, affects the algorithms needed to computationally process the languages, hence, the prospects for bootstrapping.

Note that middle cluster looks deceptively isolating, but it isn’t. Sesotho and Setswana are statistically significantly different from the others, in that they are even more disjunctive than English. Sepedi (top-most line) even more so. While I don’t know that language, a hypothetical example suffice to illustrate this notion. There is conjugation of verbs, like ‘works’ or trabajas or usebenza (inflection underlined), but some orthographer a while ago could have decided to write that separate from the verb stem (e.g., trabaj as and u sebenza instead), hence, generating more tokens with fewer characters.

There are other aspects of language and orthography one can ‘play’ with to analyse quantitatively, like whether words mainly end in a vowel or not, and which vowel mostly, and whether two successive vowels are acceptable for a language (for some, it isn’t). This is further described in the paper [3].

Yet, the UDHR is just one document. To examine the generalisability of these observations, we need to know whether the UDHR text is a ‘typical’ one. This was assessed in more detail by zooming in on isiZulu both quantitatively and qualitatively with four other corpora and texts in different genres. The results show that the UHDR is a typical text document orthographically, at least for the cumulative frequency distribution of the word length.

There were some other differences across the other corpora, which have to do with genre and datedness, which was observed elsewhere for whole words [4]. For instance, news items of isiZulu newspapers nowadays include words like iFacebook and EFF, which surely don’t occur in a century-old bible translation. They do violate the ‘no two successive vowels’ rule and the ‘final vowel’ rule, though.

On the qualitative side of the matter, and which will have an effect on searching for information in texts, text summarization, and error correction of spellcheckers, is, again, that agglutination. For instance, searching on imali ‘money’ alone would be woefully inadequate to find all relevant texts; e.g., those news items also include kwemali, yimali, onemali, osozimali, kwezimali, and ngezimali, which are, respectively of -, and -, that/which/who has -, of – (pl.), about/by/with/per – (pl.) money. Searching on the stem or root only is not going to help you much either, however. Take, for instance -fund-, of which the results of just two days of Isolezwe news articles is shown in the table below (articles from 2015, when there were protests, too). Depending on what comes before fund and what comes after it, it can have a different meaning, such as abafundi ‘students’ and azifundi ‘they do not learn’.

Placing this is the broader NLP scope, it also affects the widely-used notion of lexical diversity, which, in its basic form, is a type-to-token ratio. Lexical diversity is used as a proxy measure for ‘difficulty’ or level of a text (the higher the more difficult), language development in humans as they grow up, second-language learning, and related topics. Letting that loose on isiZulu text, it will count abafundi, bafundi, and nabafundi as three different tokens, so wheehee, high lexical diversity, yet in English, it amounts to ‘students’, ‘students’ and ‘and the students’. Put differently, somehow we have to come up with a more meaningful notion of lexical diversity for agglutinating languages. A first attempt is made in the paper in its section 4 [3].

Thus, the last word has not been said yet about orthographic similarity, yet we now do have more insight into it. The surprising similarity of isiZulu (South Africa) with Runyankore (Uganda) was exploited in another research activity, and shown to be very amenable to bootstrapping [5], so, in its own way providing supporting evidence for bootstrapping potential that the figure above also indicated as promising.

As a final comment on the tooling side of things, I did use NLTK (Python). It worked well for basic analyses of text, but it (and similar NLP tools) will need considerable customization for the agglutinating languages.

 

References

[1] C. Meinhof. 1932. Introduction to the phonology of the Bantu languages . Dietrich Reiner/Ernst Vohsen, Johannesburg. Translated, revised and enlarged in collaboration with the author and Dr. Alice Werner by N.J. Van Warmelo.

[2] L. Pretorius and S. Bosch. Exploiting cross-linguistic similarities in Zulu and Xhosa computational morphology: Facing the challenge of a disjunctive orthography. In Proceedings of the EACL 2009 Workshop on Language Technologies for African Languages – AfLaT 2009, pages 96–103, 2009.

[3] C.M. Keet. An assessment of orthographic similarity measures for several African languages. Technical report, arxiv 1608.03065. August 2016.

[4] Ndaba, B., Suleman, H., Keet, C.M., Khumalo, L. The Effects of a Corpus on isiZulu Spellcheckers based on N-grams. IST-Africa 2016. May 11-13, 2016, Durban, South Africa.

[5] J. Byamugisha, C. M. Keet, and B. DeRenzi. Bootstrapping a Runyankore CNL from an isiZulu CNL. In B. Davis et al., editors, 5th Workshop on Controlled Natural Language (CNL’16), volume 9767 of LNAI, pages 25–36. Springer, 2016. 25-27 July 2016, Aberdeen, UK.