systems that know and understand and think and learn
Deep learning can produce some impressive chatbots, but they are hardly intelligent. In fact, they are precisely ignorant in that they do not think or know anything.
More intelligent dialog with an artificially intelligent agent involves both knowledge and thinking. In this article, we educate an intelligent agent that reasons to answer questions.
Benjamin Grosof, co-founder of Coherent Knowledge Systems, is also involved with developing a standard ontology for the financial services industry (i.e., FIBO). In the course of working on FIBO, he is developing a demonstration of defeasible logic concerning Regulation W of the The Federal Reserve Act. Regulation W specifies which transactions involving banks and their affiliates are prohibited under Section 23A of the Act. In the course of doing this, there are various documents which are being captured within the Linguist™ platform. This is a brief note of how those documents can be imported into the platform for curation into formal semantics and logic (as Benjamin and Coherent are doing). Continue reading “Affiliate Transactions covered by The Federal Reserve Act (Regulation W)”
As I mentioned in this post, we’re having fun layering questions and answers with explanations on top of electronic textbook content.
The basic idea is to couple a graph structure of questions, answers, and explanations into the text using semantics. The trick is to do that well and automatically enough that we can deliver effective adaptive learning support. This is analogous to the knowledge graph that users of Knewton‘s API create for their content. The difference is that we get the graph from the content, including the “assessment items” (that’s what educators call questions, among other things). Essentially, we parse the content, including the assessment items (i.e., the questions and each of their answers and explanations). The result of this parsing is, as we’ve described elsewhere, precise lexical, syntactic, semantic, and logic understanding of each sentence in the content. But we don’t have to go nearly that far to exceed the state of the art here. Continue reading “Automatic Knowledge Graphs for Assessment Items and Learning Objects”
We’re collaborating on some educational work and came across this sentence in a textbook on finance and accounting:
We use statistical NLP but assist with the ambiguities. In doing this, we relate questions and answers and explanations to the text.
We also extract the terminology and produce a rich lexicalized ontology of the subject matter for pedagogical uses, assessment, and adaptive learning.
Here’s one that just struck me as interesting. This is a case where the choice looks like it won’t matter much either way, but …
In developing a compliance application based on the institutional review board policies of John Hopkins’ Dept. of Medicine, we have to clarify the following sentence:
As you can see, there are a number of compound words and acronyms, as well as references to the Code of Federal Regulations that need to be defined or recognized to understand this sentence. Continue reading “Knowledge acquisition using lexical and semantic ontology”
In preparation for generating RIF and SBVR from the Linguist, we have produced an OWL ontology for the pertinent aspects of the SBVR specification. We hope that this is helpful to others and would sincerely appreciate any corrections or comments on how to improve it.
Paul
We’re working with the English Resource Grammar (ERG), OWL, and Vulcan’s SILK to educate the machine by translating textbooks into defeasible logic. Part of this involves an ontology that models semantics more deeply than the ERG, which is based on head-driven phrase structure grammar (HPSG), which provides deeper parsing and, with the ERG and the DELPH-IN infrastructure, also provides a simple under-specified semantic representation called minimal recursion semantics (MRS).
We’re having a great time using OWL to clarify and enrich the semantics of the rich model underlying the ERG. Here’s an example, FYI. If you’d like to know more (or help), please drop us a line! Overall the project will demonstrate our capabilities for transforming everyday sentences into RIF and business rule languages using SBVR extended with defeasibility and other capabilities, all modeled in the same OWL ontology.
What triggered this blog entry was a bit of a surprise in seeing that whether or not an adjective could be used depictively is sometimes encoded in the lexicon. This is one of the problems of TDL versus a description-logic based model with more expressiveness. It results in more lexical entries than necessary, which has been discussed by others when contrasted with the attributed logic engine (ALE), for example.
In trying to model the semantics of words like ‘same’ and ‘different’, we are scratching our heads about these lines from the ERG’s lexicon:
One of the interesting things about lexicalized grammars is that lexical entries (i.e., ‘words’) are described with almost arbitrary combinations of their lexical, syntactic, and semantic characteristics.
The preceding code is expressed in a type description language (TDL) used by the Lisp-based LKB (and its C++ counterpart, PET, which are unification-based parsers that produce a chart of plausible parses with some efficiency. What is given above is already deeper than what you can expect from a statistical parser (but richer descriptions of lexical entries promises to make statistical parsing much better, too).
Unfortunately, there is no available documentation on why the ERG was designed as it is, so the meaning of the above is difficult to interpret. For example, the types of lexical entries (the symbols ending in ‘_le’) referenced above are defined as follows:
Needless to say, that’s a mouthful! Chasing this down, the following ‘informs’ us that “the same”, which uses type #2 above, is defined using the following lexical types:
Which is to say that it is non-conjunctive, complements a head to form a phrase, can’t be affixed, cannot constitute a main clause, and is a word.
The fact that the lexical entry for “the same” is adjectival is given the definition of the following type(s) used in the SYNSEM feature:
Which is to say that it is a comparative or superlative adjectival word (even though it consists of two lexemes in its ‘orthography’) that involves two semantic arguments including one complement which may be unexpressed prepositional phrase. A comparative or superlative adjective, in turn, is non-conjunctive, complements a head to form a phrase, is non-affix bearing (?), and non-clausal, as defined by the type ‘nonc-hm-nab’ above.
The types used in the syntax and semantic (i.e., SYNSEM) feature of the two lexical types are defined as follows (none of which is documented):
In a moment, we’ll discuss the types used in the second of these, but first, some basics on the semantics that are mixed with the syntax above.
In effect, the above indicates that a new ‘elementary predication’ will be needed in the MRS to represent the adjectival relationship in the logic derived in the course of parsing (i.e., that’s what ‘unsp_ind’ means, although it’s not documented, which I will try not to bemoan much further.)
The following indicates that the newly formed elementary predicate is not (initially) within any scope and that it has two arguments whose semantics (i.e., their RELations) are concatenated for propagation into the list of elementary predications that will constitute the MRS for any parses found.
The last of these expresses that the constuction is lexical rather than phrasal (which includes clausal in the ERG).
Continuing with the definition of “the same” as an adjective, the following finally clarifies what it means to be a basic adjective:
Well, ‘clarifies’ might not have been the right word! Essentially, it indicates that the adjective may have an optional degree specifier (which semantically modifies the predicate of the adjective) and that the predicate specified in the lexical entry becomes the predicate used in the MRS. The rest is defined below:
Which ends with a bunch of basic setup types except for constraining for relation for an adjective to be ‘basically adjectivally’ on the first two lines. Also on these first two lines, it specifies that its subject and its specifier, if any, must be completed (i.e., empty) and agree with its non-conjunctive argument (which is not to say that it cannot be conjunctive, but that it modifies the conjunction as a whole, if so.) Whether or not it is expressed will determine if there are any further predicates about its arguments or if its unexpressed argument is identified by an otherwise unreferenced variable in any resulting MRS.
The lexical grounding of this type specification is given below, indicating that it may (or not) have phonology (e.g., pronunciation, such as whether its onset is voiced) and if and how and with what punctuation it may appear, if any. In general, a semantic argument may be lexical or phrasal and optional but if it appears it corresponds to some semantic index (think variable) in sort of predicate in any resulting MRS. (The *_min types do not constrain the values of their features any further).
Note that an adjective is not possess-able and that it modifies something nominal (or a title) and that if it has a specifier that it is a quantifier or degree (e.g., ‘very’). Again, an adjective cannot function as a main clause or be conjunctive (in and of itself).
Finally, if you look far above you will see that the basic semantics of an adjective with an additional semantic argument is ‘intersective’, as in:
Here, the length 0 difference list and the following definitions indicate that intersective semantics do not accept anything but local modification:
AVM stands for ‘attribute value matrix’, which is the structure by which types and their features are defined (with nesting and unification constraints using # to indicate equality).
By now you’re probably getting the idea that there is fairly significant model of the English language, including its lexical and syntactic aspects, but if you look there is a lot about semantics here, too.
The slides for my Business Rules Forum presentation on event semantics and focusing on events in order to simplify process definition and to facilitate more robust governance and compliance are at Event-centric BPM.
After the talk I spoke with Jan Verbeek and Gartjan Grijzen of Be Informed and reviewed their software, which is excellent. They have been quite successful with various government agencies in applying the event-centric methodology to produce goal-driven processing. Their approach is elegant and effective. It clearly demonstrates the merits of an event-centric approach and the power that emerges from understanding event-dependencies. Also, it is very semantic, ontological, and logic-programming oriented in its approach (e.g., they use OWL and a backward-chaining inference engine).
They do not have the top-down knowledge management approach that I advocate nor do they provide the logical verification of governing policies and compliance (i.e., using theorem provers) that I mention in the talk (see Guido Governatori‘s 2010 publications and Travis Breaux‘s research at CMU, for example) but theirs is the best commercially deployed work in separating business process description from procedural implementation that comes to mind. (Note that Ed Barkmeyer of NIST reports some use of SBVR descriptions of manufacturing processes with theorem provers. Some in automotive and aerospace industries have been interested in this approach for quality purposes, too.)
BeInformed is now expanding into the United States with the assistance of Mills Davis and others. Their software is definitely worth consideration and, in my opinion, is more elegant and effective than the generic BPMN approach.
The standard for defining ontologies these days is OWL and Protege. Unfortunately, OWL lacks any notion of exceptions in inheritance or any other notion of defeasibility.
So, although you may want to say that birds fly, you’re ontology will be broken (or become much more complicated) when you realize there are birds that can’t fly, such as penguins or ostriches, or even sick or injured birds.
Practically speaking, you need something like courteous logic or the defeasibility in SILK to handle this (or any 1980s expert system shell or even earlier frame system). OWL is very hard on mortal man (e.g., mainstream IT) in this regard.
How can I tell OWL that a pronoun is a noun but that pronouns are a closed class of words, unlike nouns, verbs, adjectives, and adverbs (in general). Well, I’ll have to tell it about open-class nouns versus closed class nouns. What a pain!
This is why we use Protege primarily as a drafting tool and, for example, SILK, to do reasoning. Non-defeasible description logic and first-order reasoners are difficult to get along with, in practice (and make sustainable knowledge repositories too difficult – which inhibits adoption, obviously).
The basic formal ontology (BFO) offers a simple, elegant process model. It adds alethic and teleological semantics to the more procedural models, among which I would include NIST’s process specification language (PSL) along with BPMN.
Although alethic typically refers to necessary vs. possible, it clearly subsumes the probable or expected (albeit excluding deontics0). For example, consider the notion of ‘disposition’ (shown below as rendered in Protege):
For example, cells might be disposed to undergo the cell cycle, which consists of interphase, mitosis, and cytokinesis. Iron is disposed to rust. Certain customers might be disposed to comment, complain, or inquire.
Disposition is nice because it reflects things that have an unexpected high probability of occurring1 but that may not be a necessary part of a process. It seems, however, that disposition is lacking from most business process models. It is prevalent in the soft and hard sciences, though. And it is important in medicine.
Disposition is distinct from what should occur or be attempted next in a process. Just because something is disposed to happen does not mean that it should or will. Although disposition is clearly related to business events and processes, it seems surprisingly lacking from business models (and CEP/BPM tooling).2
A teleological aspect of BFO is the notion of purpose or intended ‘function’, as shown below:
Function is about what something is expected to do or what it is for. For example, what is the function of an actuary? Representing such functionality of individuals or departments within enterprises may be atypical today, but is clearly relevant to skills-based routing, human resource optimization and business modeling in general.
Understanding disposition and function is clearly relevant to business modeling (including organizational structure), planning and performance optimization. Without an understanding of disposition, anticipation and foresight will be lacking. Without an understanding of function, measurement, reporting, and performance improvement will be lacking.
1 Thanks to BG for “politicians are disposed to corruption” which indicates a population that is more likely than a larger population to be involved in certain situations.
2 Cyc’s notion of ‘disposition’ or ‘tendency’ is focused on properties rather than probabilities, as in the following citation from OpenCyc. Such a notion is similarly lacking from most business models, probably because its utility requires more significant reasoning and business intelligence than is common within enterprises.
The collection of all the different quantities of dispositional properties; e.g. a particular degree of thermal conductivity. The various specializations of this collection are the collections of all the degrees of a particular dispositional property. For example, ThermalConductivity is a specialization of this collection and its instances are usually denoted with the generic value functions as in (HighAmountFn ThermalConductivity).
