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.

More stuff: relating stuffs and amounts of stuff to their parts and portions

With all the protests going on in South Africa, writing this post is going to be a moment of detachment of it (well, I’m trying), for it concerns foundational aspects of ontologies with respect to “stuff”. Stuff is the philosophers’ funny term for those kind of things that cannot be counted, or only counted in quantities, and are in natural language generally referred to by mass nouns. For instance, water, gold, mayonnaise, oil, and wine as kinds of things, yet one can talk of individual objects of them only in quantities, like a glass of wine, a spoonful of mayonnaise, and a litre of oil. It is one thing to be able to say which types of stuff there are [1], it is another matter how they relate to each other. The latter is described in the paper recently accepted at the 20th International Conference on Knowledge Engineering and Knowledge management (EKAW’16), entitled “Relating some stuff to other stuff” [2].

Is something like that even relevant, when students are protesting for free education, among other demands? Yes. At the end of the day, it is part and parcel of a healthy environment to live in. For instance, one should be able to realise traceability in food and medicine supply chains, to foster practices, and check compliance, of good production processes and supply chains, so that you will not buy food that makes you ill or take medicines that are fake [3,4]. Such production processes and product logistics deal with ‘stuffs’ and their portions and parts that get separated and put together to make the final product. Current implementations have only underspecified ‘links’ (if at all) that doesn’t let one infer automatically what (or who) the culprit is. Existing theoretical accounts from philosophy and in domain ontologies are incomplete, so they wouldn’t help you further either. The research described in the paper solves this issue.

Seven relations for portions and stuff-parts were identified, which have a temporal dimension where needed. For instance, the upper-half of the wine in your wine glass is a portion of the whole amount of wine in the glass, yet that amount of wine was a portion of the amount of wine in the bottle when you opened it, and yet it has as part some amount of alcohol. (Some reader may not find this example nice, for it being with alcohol, but Western Cape, where Cape Town is situated, is the wine region of the country.) The relations are structured in a little hierarchy, as informally depicted in the figure below.

Section of the basic taxonomy of part-whole relations of [5] (less and irrelevant sections in grey or suppressed), extended with the stuff relations and their position in the hierarchy.

Section of the basic taxonomy of part-whole relations of [5] (less and irrelevant sections in grey or suppressed), extended with the stuff relations and their position in the hierarchy.

Their formal definitions are included in the paper.

Another aspect of the solution is that it distinguishes between 1) the extensional and intensional level—like, between ‘an amount of wine’ and ‘wine’—because different constraints apply (following from that latter can be instantiated the former cannot), and 2) the amount of stuff and the (repeatable) quantity, as one can have 1kg of many things.

Just theory isn’t good enough, though, for one would want to use it in some way to indeed get those benefits of traceability in the supply chains. After considering the implementation options (see paper for details), I settled for an extension to the Stuff Ontology core ontology that now also imports a special purpose module OMmini of the Ontology of Units of Measure (see also the Stuff Ontology page). The latter sounds easier than that it worked in praxis, but that’s a topic of a different post. The module is there, and the links between the OMmin.owl and stuff.owl have been declared.

Although the implementation is atemporal in the end, it is still possible to do some automated reasoning for traceability. This is mainly thought availing of property chains to approximate the relevant temporal aspects. For instance, with scatteredPortionOf \circ portionOf \sqsubseteq scatteredPortionOf then one can infer that a scattered portion in my glass of wine that was a portion of bottle #1234 of organic Pinotage wine of an amount of wine, contained in cask #3, with wine from wine farm X of Stellar Winery from the 2015 harvest is a scattered portion of that amount of matter (that cask). Or take the (high-level) pharmaceutical supply chain from [4]: a portion (that is on a ‘pallet’) of the quantity of medicine produced by the manufacturer goes to the warehouse, of which a portion (in a ‘case’) goes to the distribution centre. From there, a portion ends up on the dispensing shelf, and someone buys it. Then tracing any customer’s portion of medicine—i.e., regardless the actual instance—can be inferred with the following chain: scatteredPortionOf \circ scatteredPortionOf \circ scatteredPortionOf \sqsubseteq scatteredPortionOf

Sure, the research presented hasn’t solved everything yet, but at least software developers now have a (better) way to automate traceability in supply chains. It also allows one to be more fine-grained in the analysis where a culprit may be, so that there are fewer cases of needless scares. For instance, we know that when there’s an outbreak of Salmonella, then we only have to trace where the batch of egg yolk went (typically in the tiramisu served in homes for the elderly), where it came from (which farm), and got mixed with in the production process, while the amounts of egg white on your lemon merengue still would be safe to eat even when it came from the same batch that had at least one infected egg.

I’ll be presenting the paper at EKAW’16 in November in Bologna, Italy, and hope to see you there! It’s not a good time of the year w.r.t. weather, but that’s counterbalanced by the beauty of the buildings and art works, and the actual venue room is in one of the historical buildings of the oldest university of Europe.

 

References

[1] Keet, C.M. A core ontology of macroscopic stuff. 19th International Conference on Knowledge Engineering and Knowledge Management (EKAW’14). K. Janowicz et al. (Eds.). 24-28 Nov, 2014, Linkoping, Sweden. Springer LNAI vol. 8876, 209-224.

[2] Keet, C.M. Relating some stuff to other stuff. 20th International Conference on Knowledge Engineering and Knowledge Management EKAW’16). Springer LNAI, 19-23 November 2016, Bologna, Italy. (accepted)

[3] Donnelly, K.A.M. A short communication – meta data and semantics the industry interface: what does the food industry think are necessary elements for exchange? In: Proc. of Metadata and Semantics Research (MTSR’10). Springer CCIS vol. 108, 131-136.

[4] Solanki, M., Brewster, C. OntoPedigree: Modelling pedigrees for traceability in supply chains. Semantic Web Journal, 2016, 7(5), 483-491.

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

On generating isiZulu sentences with part-whole relations

It all sounded so easy… We have a pretty good and stable idea about part-whole relations and their properties (see, e.g., [1]), we know how to ‘verbalise’/generate a natural language sentence from basic description logic axioms with object properties that use simple verbs [2], like Professor \sqsubseteq \exists teaches.Course ‘each professor teaches at least one course’, and SNOMED CT is full of logically ‘simple’ axioms (it’s in OWL 2 EL, after all) and has lots of part-whole relations. So why not combine that? We did, but it took some more time than initially anticipated. The outcomes are described in the paper “On the verbalization patterns of part-whole relations in isiZulu”, which was recently accepted at the 9th International Natural Language Generation Conference (INLG’16) that will be held 6-8 September in Edinburgh, Scotland.

What it ended up to be, is that notions of ‘part’ in isiZulu are at times less precise and other times more precise compared to the taxonomy of part-whole relations. This interfered with devising the sentence generation patterns, it pushed the number of ‘elements’ to deal with in the language up to 13 constituents, and there was no way to avoid proper phonological conditioning. We already could handle quantitative, relative, and subject concords, the copulative, and conjunction, but what had to be added were, in particular, the possessive concord, locative affixes, a preposition (just the nga in this context), epenthetic, and the passive tense (with modified final vowel). As practically every element has to be ‘completed’ based on the context (notably the noun class), one can’t really speak of a template-based approach anymore, but a bunch of patterns and partial grammar engine instead. For instance, plain parthood, structural parthood, involvement, membership all have:

  • (‘each whole has some part’) QCall_{nc_{x,pl}} W_{nc_{x,pl}} SC_{nc_{x,pl}}-CONJ-P_{nc_y} RC_{nc_y}-QC_{nc_y}- dwa
  • (‘each part is part of some whole’) QCall_{nc_{x,pl}} P_{nc_{x,pl}} SC_{nc_{x,pl}}-COP-ingxenye PC_{\mbox{\em ingxenye}}-W_{nc_y} RC_{nc_y}-QC _{nc_y}- dwa

There are a couple of noteworthy things here. First, the whole-part relation does not have one single string, like a ‘has part’ in English, but it is composed of the subject concord (SC) for the noun class (nc) of the noun that play the role of the whole ( W ) together with the phonologically conditioned conjunction na- ‘and’ (the “SC-CONJ”, above) and glued onto the noun of the entity that play the role of the part (P). Thus, the surface realisation of what is conceptually ‘has part’ is dependent on both the noun class of the whole (as the SC is) and on the first letter of the name of the part (e.g., na-+i-=ne-). The ‘is part of’ reading direction is made up of ingxenye ‘part’, which is a noun that is preceded with the copula (COP) y– and together then amounts to ‘is part’. The ‘of’ of the ‘is part of’ is handled by the possessive concord (PC) of ingxenye, and with ingxenye being in noun class 9, the PC is ya-. This ya- is then made into one word together with the noun for the object that plays the role of the whole, taking into account vowel coalescence (e.g., ya-+u-=yo-). Let’s illustrate this with heart (inhliziyo, nc9) standing in a part-whole relation to human (umuntu, NC1), with the ‘has part’ and ‘is part of’ underlined:

  • bonke abantu banenhliziyo eyodwa ‘All humans have as part at least one heart’
    • The algorithm, in short, to get this sentence from, say Human \sqsubseteq \exists hasPart.Heart : 1) it looks up the noun class of umuntu (nc1); 2) it pluralises umuntu into abantu (nc2); 3) it looks up the quantitative concord for universal quantification (QCall) for nc2 (bonke); 4) it looks up the SC for nc2 (ba); 5) then it uses the phonological conditioning rules to add na- to the part inhliziyo, resulting in nenhliziyo and strings it together with the subject concord to banenhliziyo; 6) and finally it looks up the noun class of inhliziyo, which is nc9, and from that it looks up the relative concord (RC) for nc9 (e-) and the quantitative concord for existential quantification (QC) for nc9 (being yo-), and strings it together with –dwa to eyodwa.
  • zonke izinhliziyo ziyingxenye yomuntu oyedwa ‘All hearts are part of at least one human’
    • The algorithm, in short, to get this sentence from Heart \sqsubseteq \exists isPartOf.Human : 1) it looks up the noun class of inhliziyo (nc9); 2) it pluralises inhliziyo to izinhliziyo (nc10); 3) it looks up the QCall for nc10 (zonke); 4) it looks up the SC for nc10 (zi-), takes y- (the COP) and adds them to ingxenye to form ziyingxenye; 5) then it uses the phonological conditioning rules to add ya- to the whole umuntu, resulting in yomuntu; 6) and finally it looks up the noun class of umuntu, which is nc1, and from that the RC for nc10 (o-) and the QC for nc10 (being ye-), and strings it together with –dwa to oyedwa.

For subquantities, we end up with three variants: one for stuff-parts (as in ‘urine has part water’, still with ingxenye for ‘part’), one for portions of solid objects (as in ‘tissue sample is a subquantity of tissue’ or a slice of the cake) that uses umunxa instead of ingxenye, and one ‘spatial’ notion of portion, like that an operating theatre is a portion of a hospital, or the area of the kitchen where the kitchen utensils are is a portion of the kitchen, which uses isiqephu instead of ingxenye. Umunxa is in nc3, so the PC is wa- so that with, e.g., isbhedlela ‘hospital’ it becomes wesibhedlela ‘of the hospital’, and the COP is ng- instead of y-, because umunxa starts with an u. And yet again part-whole relations use locatives (like the containment type of part-whole relation). The paper has all those sentence generation patterns, examples for each, and explanations for them.

The meronymic part-whole relations participation and constitution have added aspects for the verb, such as generating the passive for ‘constituted of’: –akha is ‘to build’ for objects that are made/constituted of some matter in some structural sense, else –enza is used. They are both ‘irregular’ in the sense that it is uncommon that a verb stem starts with a vowel, so this means additional vowel processing (called hiatus resolution in this case) to put the SC together with the verb stem. Then, for instance za+akhiwe=zakhiwe but u+akhiwe=yakhiwe (see rules in paper).

Finally, this was not just a theoretical exercise, but it also has been implemented. I’ll readily admit that the Python code isn’t beautiful and can do with some refactoring, but it does the job. We gave it 42 test cases, of which 38 were answered correctly; the remaining errors were due to an ‘incomplete’ (and unresolvable case for any?) pluraliser and that we don’t know how to systematically encode when to pick akha and when enza, for that requires some more semantics of the nouns. Here is a screenshot with some examples:

examples2

The ‘wp’ ones are that a whole has some part, and the ‘pw’ ones that the part is part of the whole and, in terms of the type of axiom that each function verbalises, they are of the so-called ‘all some’ pattern.

The source code, additional files, and the (slightly annotated) test sentences are available from the GENI project’s website. If you want to test it with other nouns, please check whether the noun is already in nncPairs.txt; if not, you can add it, and then invoke the function again. (This remains this ‘clumsily’ until we make a softcopy of all isiZulu nouns with their noun classes. Without the noun class explicitly given, the automatic detection of the noun class is not, and cannot be, more than about 50%, but with noun class information, we can get it up to 90-100% correct in the pluralisation step of the sentence generation [4].)

 

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. Basics for a grammar engine to verbalize logical theories in isiZulu. 8th International Web Rule Symposium (RuleML’14), A. Bikakis et al. (Eds.). Springer LNCS vol. 8620, 216-225. August 18-20, 2014, Prague, Czech Republic.

[3] 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. (in print)

[4] 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. (in print)

Automatically finding the feasible object property

Late last month I wrote about the updated taxonomy of part-whole relations and claimed it wasn’t such a big deal during the modeling process to have that many relations to choose from. Here I’ll back up that claim. Primarily, it is thanks to the ‘Foundational Ontology and Reasoner enhanced axiomatiZAtion’ (FORZA) approach which includes the Guided ENtity reuse and class Expression geneRATOR (GENERATOR) method that was implemented in the OntoPartS-2 tool [1]. The general idea of the GENERATOR method is depicted in the figure below, which outlines two scenarios: one in which the experts perform the authoring of their domain ontology with the help of a foundational ontology, and the other one without a foundational ontology.

generator

I think the pictures are clearer than the following text, but some prefer text, so here goes the explanation attempt. Let’s start with scenario A on the left-hand side of the figure: a modeller has a domain ontology and a foundational ontology and she wants to relate class two domain classes (indicated with C and D) and thus needs to select some object property. The first step is, indeed, selecting C and D (e.g., Human and Heart in an anatomy ontology); this is step (1) in the Figure.

Then (step 2) there are those long red arrows, which indicate that somehow there has to be a way to deal with the alignment of Human and of Heart to the relevant categories in the foundational ontology. This ‘somehow’ can be either of the following three options: (i) the domain ontology was already aligned to the foundational ontology, so that step (2) is executed automatically in the background and the modeler need not to worry, (ii) she manually carries out the alignment (assuming she knows the foundational ontology well enough), or, more likely, (iii) she chooses to be guided by a decision diagram that is specific to the selected foundational ontology. In case of option (ii) or (iii), she can choose to save it permanently or just use it for the duration of the application of the method. Step (3) is an automated process that moves up in the taxonomy to find the possible object properties. Here is where an automated reasoner comes into the equation, which can step-wise retrieve the parent class, en passant relying on taxonomic classification that offers the most up-to-date class hierarchy (i.e., including implicit subsumptions) and therewith avoiding spurious candidates. From a modeller’s viewpoint, one thus only has to select which classes to relate, and, optionally, align the ontology, so that the software will do the rest, as each time it finds a domain and range axiom of a relationship in which the parents of C and D participate, it is marked as a candidate property to be used in the class expression. Finally, the candidate object properties are returned to the user (step 4).

While the figure shows only one foundational ontology, one equally well can use a separate relation ontology, like PW or PWMT, which is just an implementation variant of scenario A: the relation ontology is also traversed upwards and on each iteration, the base ontology class is matched against relational ontology to find relations where the (parent of the) class is defined in a domain and range axiom, also until the top is reached before returning candidate relations.

The second scenario with a domain ontology only is a simplified version of option A, where the alignment step is omitted. In Figure-B above, GENERATOR would return object properties W and R as options to choose from, which, when used, would not generate an inconsistency (in this part of the ontology, at least). Without this guidance, a modeler could, erroneously, select, say, object property S, which, if the branches are disjoint, would result in an inconsistency, and if not declared disjoint, move class C from the left-hand branch to the one in the middle, which may be an undesirable deduction.

For the Heart and Human example, these entities are, in DOLCE terminology, physical objects, so that it will return structural parthood or plain parthood, if the PW ontology is used as well. If, on the other hand, say, Vase and Clay would have been the classes selected from the domain ontology, then a constitution relation would be proposed (be this with DOLCE, PW, or, say, GFO), for Vase is a physical object and Clay an amount of matter. Or with Limpopo and South Africa, a tangential proper parthood would be proposed, because they are both geographic entities.

The approach without the reasoner and without the foundational ontology decision diagram was tested with users, and showed that such a tool (OntoPartS) made the ontology authoring more efficient and accurate [2], and that aligning to DOLCE was the main hurdle for not seeing even more impressive differences. This is addressed with OntoPartS-2, so it ought to work better. What still remains to be done, admittedly, is that larger usability study with the updated version OntoPartS-2. In the meantime: if you use it, please let us know your opinion.

 

References

[1] Keet, C.M., Khan, M.T., Ghidini, C. Ontology Authoring with FORZA. 22nd ACM International Conference on Information and Knowledge Management (CIKM’13). ACM proceedings, pp569-578. Oct. 27 – Nov. 1, 2013, San Francisco, USA.

[2] Keet, C.M., Fernandez-Reyes, F.C., Morales-Gonzalez, A. Representing mereotopological relations in OWL ontologies with OntoPartS. 9th Extended Semantic Web Conference (ESWC’12), Simperl et al. (eds.), 27-31 May 2012, Heraklion, Crete, Greece. Springer, LNCS 7295, 240-254.

FAIR’14 and ‘modelling relationships’ tutorial

After a weekend of ‘loadshedding’ (one of those South African euphemisms) I’m posting a few notes on the Forum on Artificial Intelligence Research 2014 (FAIR’14) that took place from 3-5 Dec 2014 at Stellenbosch University, which was organised by CAIR and co-located with the FASTAR/Espresso Workshop 2014, which, in turn, was co-located with PRASA, AFLaT, and RobMech 2014 in Cape Town. FAIR’14 consisted of a presentation by Sergei Obiedkov of the Higher School of Economics, Russia, a tutorial on modelling relationships in ontologies by me, and a course on computational social choice theory by Ulle Endriss from the ILLC, University of Amsterdam, The Netherlands.

While not quite relevant to my current research except for judgement aggregation at the end (for crowdsourcing), Ulle’s course was one of those events that made me think “[why didn’t/if only] I was exposed to this material before?!”, when I had to make choices as to what to study and specialise in (though, admitted, once knowing about the math with game theory and applying that to peace negotiations in my MA pdf, I still went on in CS with KR&R and ontologies). Ulle’s course combined socially relevant topics, such as the fair allocation of resources and voting systems, with solid, precise, logic- and math-based representations and computation. Besides the engaging content, he’s also good at teaching it. The content and slides are a condensed version of his MSc course on social choice theory and are available online here, which also has links to related reading material.

I tried to condense into 2 hours some aspects of modelling relationships in ontologies. It started with some problems and questions, proceeded to touching upon the nature of relations and some detail of the formal semantics, common relationships (with some detail about mereotopology), and closing with some practical modelling guidance and reasoner performance when modelling it one way or another. It being a tutorial, and not all participants had Protégé installed, I resorted to a peer instruction audience response system to incorporate interactively some questions about modelling some relationships. The slides are available online (though also here the text on the slides only partially reflect what I’ve talked about).

Other than that, there’s always the social component. Despite the weird time-warp that Stellenbosch town constitutes, it was really nice to catch up with former colleagues and to see the progress of postgrads of UKZN, to hear about the future of CAIR, and that it’s a small world even when meeting people new to me. And the food & wine was delicious. The train travel back to Cape Town took a bit longer than the schedule said it ought to be, but I recommend it nevertheless.

A few notes on a successful ESWC’12 and OWLED’12

Slightly later than near-realtime due to flight delays, here are a few notes on the 9th Extended Semantic Web Conference ESWC’12 and OWL: Experiences and Directions OWLED’12, which I attended about two weeks ago in Crete, Greece.

ESWC’12

ESWC’12 was as selective as previous years, with, on average, a 25% acceptance rate. The proceedings are published by Springer; where applicable, I’ve linked the freely available versions in the references below. There’s also metadata and a list of award winners.

Main background picture of the ESWC’12 conference, with Cretan hills

Keynotes

I assume that, like last year, The keynotes have been put on the video lectures website; below follows a brief impression. for now, you’ll have to make do with a brief impression through my lenses.

Alon Halevy, head of structured data at Google, gave his keynote the morning after the social dinner (but the conference hall was full nevertheless). He entertains the perspective of Knowledge Representation and the Semantic Web as being “databases on steroids”. The talk’s topics were Google fusion tables with lightweight semantics that are intended as a “data management for the 99%” and Webtables, which was about a search for data tables on the Web, with as goal to have an easy to use database system that is integrated with the web. The work on web tables was alike a very large-scale attempt at bottom-up lightweight conceptual data model and ontology development. They crawled the Web for raw tables (14 billion), of which an estimated 154 million can pass for real relations (relations from the database viewpoint, with structured data, not using a html table for the layout of a page), which then ended up as 2.5 million schemas as recovered table/relation semantics. And then there’s Halevy’s enthusiasm about coffee.

Aleksander Kolcz from Twitter went over a few problems they are trying to solve at Twitter, such as the tweet relevance, who to follow, content recommendation, language, anti-spam, and user interest modeling. As small tidbit of data: there are 140 million users, 340 million tweets/day, and 2.3 billion search queries/day (i.e., 26K/sec.). Apparently, when one has enough, i.e., very large amounts, of data, simple models work “remarkably well” and ensembles of classifiers perform better in accuracy.

Abraham Bernstein’s keynote was about getting our act together in the semantic web research area and promoting the “garbage can theory” that was introduced by Cohen, March and Olsen in 1973: or, some ideas, theories, and tools are ‘thrown away’ into the garbage, where they can meet others, and combine so that something beautiful can come of it after all (this is my simplistic, shorthand version of it).

Unfortunately I missed the pre-conference keynote by Julius van der Laar because OWLED was still ongoing. By hearsay, I’ve heard it was a good/interesting one about what (sneaky) social media strategies the Obama campaign used in the previous presidential elections in 2008.

Papers

There were several tracks that ran in parallel, hence attendance was necessarily limited due to those logistic constraints. I’ve attended the ontologies, reasoning, semantic data management, digital libraries and cultural heritage, and in use sessions. The following pointers are based on my attendance of the presentations and partial reading of the papers.

Ontologies track. Yves Raimond from the BBC presented a query-driven evaluation framework for ontologies, defining their way of ‘good’ with respect to the task and data, and applied it to the music ontology (online slides), noting some room for improvements. The paper also has a neat brief overview of techniques for ontology evaluation [1]. I presented the paper co-authored with Francis Fernandez and Annette Morales on mereotopology and the OntoPartS tool that helps modellers to represent part-whole relations [2], which I introduced in an earlier post. OntoPartS was also presented at the demo session [3], which generated quite some interest among logicians and practitioners alike. Besides my ‘toy ontology’ examples to demonstrate the tool’s functionality, Martin Hepp had brought his GoodRelations ontology for e-commerce, which I thus used instead to illustrate adding part-whole relations to a real ontology. The demo session ended officially at 9pm, but it was after 10pm before I packed up my tablet.

Semantic data management track. Craig Knoblock and co-authors developed a system to link data to ontologies and preserve the linking in a so-called (logic-based) “source model” that is computed semi-automatically by taking as input the data, an ontology, some learned semantic types, and a refinement step by the user in a nice GUI [4]. This was evaluated with a set of bio-informatics resources, such as UniProt. The presentation by Lorena Etcheverry was a bit long on the intro, but the idea nice: enhancing OLAP analysis with ‘good enough’ temporary cubes generated from web sources, the introduction of a new vocabulary, Open Cubes, for the specification and publication of multidimensional cubes on the Semantic Web (which, unfortunately, the authors still have not shared online), and an algorithm for creating the SPARQL 1.1 query for rollup [5].

In use track. Michel Dumontier demonstrated an extension to the HyQue hypothesis formulator and evaluator, using rules sets using the SPARQL Inferencing Notation (SPIN) so that users can trace their hypothesis evaluation [6]. Stefan Scheglmann presented a paper on their efforts how to provide “programming access” to ontologies and have an accompanying tool OntoMDE, a model-driven engineering toolkit (which, however, does not seem to be online available, although a link was shown in the presentation, and I jotted down something on Eclipse plugins) [7]. StorySpace was put in the Digital Libraries and cultural heritage track, but could just as well have been in in-use: it is an environment for constructing and navigating stories, plots, and narratives, guided by the newly introduced curate ontology [8]. We’ll have to look at all that in more detail in the context of our IKMS development [9].

OWLED’12

The proceedings of OWLED’12 are available on CEUR-WS. Over 30 papers were submitted, so, the workshop ended up to be somewhat selective compared to previous years. 18 papers were presented, a keynote, and two tutorials. The following is, again, a selection of that (mainly due to my time constraints reading the papers and typing up something).

Mariano Rodriguez presented the ontopQuest system [10] for Ontology-Based Data Access, providing SPARQL query answering with OWL 2 QL/RDFS entailments.  It works with the so-called “classic ABox mode” with an internal relational database and in “virtual ABox mode”, and, unlike, say, QuOnto, it embeds most of the TBox semantics into the database by availing of a (also recently developed) semantic indexing technique. (Hopefully that’ll help my ontologies & knowledge bases students to answer the OBDA questions better next time, who ought to have read at least David Toman’s slides on the principal approaches to realize OBDA before the test.) Staying with reasoning, Dmitry Tsarkov presented the idea of using metareasoning that takes into account both the features of current reasoners and modularisation to come up with the ‘best’ reasoning strategy to answer a query over only that part of the ontology that is relevant for the query [11].

An extension to the OWLGrEd tool for modeling OWL ontologies through a UML-like interface was presented: the developers have added a ‘splitter’ to enable a user to decide which axioms to close (using the OWL + Integrity Constraints), then to send the serialization to the reasoner and display the inferences [12]. Pity that it works only with the commercial RDF database Stardog by Clark & Parsia. Bijan Parsia  presented—among other things—a paper on automatically generating analogy questions, which are widely used in multiple choice questions, and determining somehow their difficulty. The automated generation was facilitated by an ontology, and the initial results are promising [13]. I presented the paper on OWL requirements for indigenous knowledge management systems [9], about which I blogged earlier, as one of my co-authors, Ronell Alberts, was already presenting a paper based on her recently completed MSc thesis [14].

One of the tutorials was about modularity, which was presented by Chiara del Vescovo and Dmitry Tsarkov from Manchester University (see their modularity website for more info). The tutorial presented an overview of where modularity is useful, and how. Some of the reasons to modularise are to facilitate the explanation services, to perform incremental reasoning, semantic diff, and hotspot detection (= splitting an ontology into the simple and the complex part). That is, it presented a viewpoint on modularity as possible solution for the issues of (and the need for) scalability and performance of automated reasoning. Modularity and modularization during modeling and to reduce the so-called cognitive overload—i.e., involving some, or even driven by, subject domain semantics—was here (and is in most other DL-oriented outlets) apparently entirely outside the scope, which is a missed opportunity (more about that another time).

Typical tourist picture of the conference hotel (the view from my room wasn’t that great, but with the busy schedule, that didn’t matter anyway)

Aside from the stimulating papers and keynotes, and ensuing conversations with fellow researchers, it was great to meet people again and meet new people, and we had a lot of fun socialising. Now back to work so as to have shot at next year’s installment of ESWC in Montpellier, France (which is close to a village I used to go on holidays for some 8 years, many years ago).

References

[1] Raimond, Y., Sandler, M. Evaluation of the music ontology framework. ESWC’12, Springer LNCS vol 7295, 255-269.

[2] Keet, C.M., Fernandez-Reyes, F.C., Morales-Gonzalez, A. Representing mereotopological relations in OWL ontologies with OntoPartS. In: Proceedings of the 9th Extended Semantic Web Conference (ESWC’12), 29-31 May 2012, Heraklion, Crete, Greece. Springer, LNCS 7295, 240-254.

[3] Morales-Gonzalez, A., Fernandez-Reyes, F.C., Keet, C.M. OntoPartS: a tool to select part-whole relations in OWL ontologies. 9th Extended Semantic Web Conference (ESWC’12), 29-31 May 2012, Heraklion, Crete, Greece. Demo with paper.

[4] Knoblock et al. Semi-automatically mapping structured sources into the semantic web. ESWC’12, Springer LNCS vol 7295, 375-390

[5] Etcheverry, L., Vaisman, A. A. Enhancing OLAP analysis with web cubes. ESWC’12, Springer LNCS vol 7295, 467-483.

[6] Callahan, A, Dumontier, M. Evaluating scientific hypotheses using the SPARQL inferecing notation. ESWC’12, Springer LNCS vol 7295, 647-658.

[7] Scheglmann, S. Scherp, A, Staab, S. Declarative Representation of Programming Access to Ontologies. ESWC’12, Springer LNCS vol 7295, 659-673.

[8] Mulholland, P., Wolff, A., and Collins, T. Curate and StorySpace: On ontology and Web-based environment for describing curatorial narrative. ESWC’12, Springer LNCS vol 7295, 748-762.

[9] Alberts, R., Fogwill, T., Keet, C.M. Several Required OWL Features for Indigenous Knowledge Management Systems. 7th Workshop on OWL: Experiences and Directions (OWLED 2012).  Klinov, P. and Horridge, M. (Eds.). 27-28 May, Heraklion, Crete, Greece. CEUR-WS Vol. 849.

[10] Rodriguez-Muro, M., Calvanese, D. Quest, an OWL 2 QL reasoner for ontology-based data access.  OWLED’12. CEUR-WS Vol. 849.

[11] Dmitry Tsarkov and Ignazio Palmisano, Divide et Impera: Metareasoning for Large Ontologies. OWLED’12. CEUR-WS Vol. 849.

[12] Kārlis Čerāns, Guntis Barzdins, Renārs Liepiņš, Jūlija Ovčiņnikova, Sergejs Rikačovs and Arturs Sprogis, Graphical Schema Editing for Stardog OWL/RDF Databases using OWLGrEd/S. OWLED’12. CEUR-WS Vol. 849.

[13] Tahani Alsubait, Bijan Parsia and Uli Sattler, Mining Ontologies for Analogy Questions: A Similarity-based Approach. OWLED’12. CEUR-WS Vol. 849.

[14] Ronell Alberts and Enrico Franconi, An integrated method using conceptual modelling to generate an ontology-based query mechanism. OWLED’12. CEUR-WS Vol. 849.

Part-whole relations, mereotopology and the OntoPartS tool

Part-whole relations are considered essential in knowledge representation and reasoning and, more practically, in ontology development and conceptual data modelling, especially in the subject domains of biology, medicine, geographic information systems, and manufacturing. In contrast to Ontology that sticks to one type of part-of, the modellers and subject domain experts have come up with a plethora of part-whole relations, some of which are considered real parthood relations and others only meronymic (or: due to imprecise natural language use). For instance, the Foundational Model of Anatomy has 8 basic locative part-whole relations [1], GALEN has come up with 26 part-whole relations [2], and in cognitive science and conceptual data modelling, it hovers around about 6 types [3,4]. They have been structured in a taxonomy of part-whole relations that makes a distinction between mereology and meronomy, transitivity and in- or non-transitivity, and the domain and range of the relationship [5], and some initial usage guidelines were proposed in [6].

But that’s not enough for the complex subject domains and demands on the representation and reasoning over the ontologies. This holds in particular when one has to represent that some things are contained in or located in something else. For instance, the way how Paris and France relate is somehow different from how the euro coin in your wallet relate to each other—the latter being an example of  (spatial) containment, but not structural part of—whereas in other case, the spatial containment of regions of space and the structural parthood of the objects occupying those regions do coincide, e.g., your heart in your body. Or consider representing that Alto Adige/Südtirol is a border province of Italy (bordering Austria), where we have to handle both the notion of administrative entities and connecting geographical regions. That is, handling regions and ‘things’ that occupy those regions (mereotopology).

Being more precise about how the things relate provides nice inferences. Take, e.g., NTPLI as ‘non-tangential proper located in’—a part is located in the whole but not at the boundary of it—and EnclosedCountry \equiv Country \sqcap \exists NTPLI.Country , with the following instances in our knowledge base NTPLI(Lesotho, South Africa) , Country(Lesotho) , and Country(South Africa) , then it deduces correctly that EnclosedCountry(Lesotho) , whereas with a mere ‘part-of’, we would not have been able to obtain this result.

Besides these examples, there are actual system requirements for, among others, annotating and querying multimedia documents and cartographic maps, such as annotating a photo of a beach where the area of the photo that depicts the sand touches the area that depicts the seawater so that, together with the knowledge that Varadero is a tangential proper part of Cuba, the semantically enhanced system can infer possible locations where the photo has been taken, or, vv., it can propose that the photo may depict a beach scene.

But how to cater for such things?

Let me summarise the three main basic problems that have to be resolved first:

  1. There is lack of oversight on plethora of part-whole relations, that include real parthood (mereology) parts with their locations (mereotopology), and other part-whole relations (from meronymy);
  2. The challenge to figure out which one to use when;
  3. The underspecified representation and reasoning consequences when one has to put up with less expressive languages for which technological infrastructure exists.

We propose to solve that in the following way, which is described in detail in [7] that recently got accepted at the 9th Extended Semantic Web Conference (ESWC’12).

The short answer for the reader who is not interested in all the theory, design, and evaluation, but just wants to model quickly: the OntoPartS tool guides you to choose the most appropriate relation and saves the selection into your OWL file.

Now for a slightly longer answer. First, we extend the taxonomy of part-whole relations of [5] with the novel addition of a taxonomy of formally defined mereotopological relations, which is driven by the KGEMT mereotoplogical theory of Varzi [8], resulting in a taxonomy of 23 part-whole relations—mereological, mereotopological, and meronymic ones—therewith ensuring a solid ontological and logic-based foundation.

Second, some things have to be simplified from the KGEMT theory to make it implementable in OWL, and we describe the design rationale and trade-offs so that OntoPartS can load OWL/OWL2-formalised ontologies, and, if desired, modify the OWL file with the chosen relation. Which OWL species is best suited obviously depends on your individual requirements, but from a representation & reasoning and mereotopology viewpoint, OWL 2 DL and OWL 2 RL seem to fit better than the other ones. (Note: there are papers on DL and representing spatial relations and on DL and parthood, and alternative representation choices are discussed in the paper, yet, as far as we are aware of, none deals with mereotopological relations in OWL or, more generally, in DL.)

Third, there is the ‘how to select’ from the 23 relations. To enable a quick selection of the appropriate relation, we avail of a simplified OWL-ized DOLCE ontology—well, just the taxonomy of categories—for the domain and range restrictions imposed on the part-whole relations and with that, we can let the user take shortcuts compared to a lengthy decision procedure. In this way, we reduced the selection procedure to 0-4 options based on just 2-3 inputs. All of this has been structured neatly in implementation-independent activity diagrams, and subsequently has been implemented; see also the demos, the tool, and the OWL version of the taxonomy of the 23 relations.

Last, we have tested OntoPartS with modellers in controlled experiments and it was shown to improve efficiency and accuracy in modeling of part-whole relations.

As mentioned, further details can be found in [7], Representing mereotopological relations in OWL ontologies with OntoPartS, which I co-authored with Francis Fernández-Reyes, with the Instituto Superior Politécnico “José Antonio Echeverría” (CUJAE), and Annette Morales-González, with the Advanced Technologies Application Center (CENATAV), both located in Cuba (the example on semantic annotation of multimedia with spatial relations comes straight from the image processing research being done at CENATAV). A tidbit of non-scientific information: the first version of the OntoPartS tool was developed as part of the mini-project that Francis, Annette (and Alexis, who is into fish fulltime now) had chosen to carry out for the ontology engineering course I taught at the University of Havana in 2010 (mentioned earlier here and here). For the paper, we added some more theory, minor refinements to the tool, and a user evaluation with several CUJAE and UKZN students and a few FUB colleagues (thanks again for their cooperation and interest). We’ve started work on additional features, so if you have any particular request, drop me a line.

References

  1. Mejino, J.L.V., Agoncillo, A.V., Rickard, K.L., Rosse, C.: Representing complexity in part-whole relationships within the foundational model of anatomy. In: Proc. of the AMIA Fall Symposium. pp. 450–454 (2003)
  2. http://www.opengalen.org/tutorials/crm/tutorial9.html up to http://www.opengalen.org/tutorials/crm/tutorial16.html/.
  3. Winston, M., Chaffin, R., Herrmann, D.: A taxonomy of part-whole relations. Cognitive Science 11(4), 417–444 (1987)
  4. Odell, J.: Advanced Object-Oriented Analysis & Design using UML. Cambridge: Cambridge University Press (1998)
  5. Keet, C.M., Artale, A.: Representing and reasoning over a taxonomy of part-whole relations. Applied Ontology 3(1-2), 91–110 (2008)
  6. Keet, C.M.: Part-whole relations in object-role models. In: Proc. of ORM’06, OTM Workshops 2006. LNCS, vol. 4278, pp. 1116–1127. Springer (2006)
  7. Keet, C.M., Fernández Reyes, F.C., Morales-González, A.: Representing mereotopological relations in OWL ontologies with OntoPartS. In Simperl, et al., eds.: Proc. of ESWC’12. LNCS, Springer (2012) 27-31 May 2012, Heraklion, Greece.
  8. Varzi, A.: Handbook of Spatial Logics, chap. Spatial reasoning and ontology: parts, wholes, and locations, pp. 945–1038. Berlin Heidelberg: Springer Verlag (2007)