1 / 29

Semantic Memory Architecture for Knowledge Acquisition and Management

Semantic Memory Architecture for Knowledge Acquisition and Management. Julian Szymański & Włodzisław Duc h Department of Electronic, Telecommunication and Informatics , Gdańsk University of Technology ; D epartment of Informatics, Nicolaus Copernicus University, Toru ń, Poland Google: Duch.

alika-tyler
Download Presentation

Semantic Memory Architecture for Knowledge Acquisition and Management

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Semantic Memory Architecture for Knowledge Acquisition and Management Julian Szymański & Włodzisław Duch Department of Electronic, Telecommunication and Informatics, Gdańsk University of Technology; Department of Informatics, Nicolaus Copernicus University, Toruń, Poland Google: Duch IMS Lhasa, 2007

  2. Motivation. Semantic memories. Talking head with memory. Implementing 20 questions game. Implementing word puzzles. Data acquisition: active learning. Future directions. Plan

  3. Motivation Language understanding is one of the key issues on the road to intelligent information management. Chatterbots are based on 40-years old template matching technology, started by ELIZA (Joseph Weizenbaum, MIT 1996). Popular: AIML language, used in Alice and many other bots. What is your favorite * ? with * = song, film, sport, team, thing, music … The “progress”: use 10.000 templates (Alice) instead of 100. Mindless cheating may get you surprisingly far … No understanding, just pre-processed answers. Sufficient for business transactions, but not for dialog. Chatterbots are quite annoying because they do not understand anything but may try to sustain dialog to kill some time. Loebner prize competitions – still poor performance.

  4. Talk to a bot … Human: There was a new earthquake in Sumatra ALICE: What happened to it? Human: Tsunami came. ALICE: I like the way you talk. Human: no ALICE: Are you serious? Human: no ALICE: Are you serious? Human: no ALICE: Are you serious? (got into the loop … ) What is wrong and what is missing? Alice has only template memory, but there are infinitely many sentences. Understanding requires concept recognition, descriptions of objects, concepts, their properties, associations, relations and possible actions in the associative semantic memory, storing basic info about the world. Dialogue: building episodic relations among concepts.

  5. Ambitious approaches… CYC, Douglas Lenat, started in 1984. Developed by CyCorp, with 2.5 millions of assertions linking over 150.000 concepts and using thousands of micro-theories (2004). Cyc-NL is still a “potential application”, knowledge representation in frames is quite complicated and thus difficult to use. Open Mind Common Sense Project (MIT): a WWW collaboration with over 14,000 authors, who contributed 710,000 sentences; used to generate ConceptNet, very large semantic network. Some interesting projects are being developed now around this network but no systematic knowledge has been collected. Other such projects: HowNet (Chinese Academy of Science), FrameNet (Berkley), various large-scale ontologies. The focus of these projects is to understand all relations in text/dialogue.

  6. Realistic goals? Different applications may require different knowledge representation. Start from the simplest knowledge representation for semantic memory. Find where such representation is sufficient, understand limitations. Drawing on such semantic memory an avatar may formulate and may answer many questions that would require exponentially large number of templates in AIML or other such language. Adding intelligence to avatars involves two major tasks: • building semantic memory model; • provide interface for natural communication. Goal: create 3D human head model, with speech synthesis & recognition, use it to interact with Web pages & local programs: a Humanized InTerface (HIT). Control HIT actions using the knowledge from its semantic memory.

  7. Web/text/databases interface Text to speech NLP functions Natural input modules Talking head Behavior control Cognitive functions Control of devices Affectivefunctions Specialized agents DREAM architecture DREAM is concentrated on the cognitive functions + real time control, we plan to adopt software from the HIT project for perception, NLP, and other functions.

  8. Types of memory Neurocognitive approach to NLP: at least 4 types of memories. Long term (LTM): recognition, semantic, episodic + working memory. Input (text, speech) pre-processed using recognition memory model to correct spelling errors, expand acronyms etc. For dialogue/text understanding episodic memory models are needed. Working memory: an active subset of semantic/episodic memory. All 3 LTM are coupled mutually providing context for recogniton. Semantic memory is a permanent storage of conceptual data. • “Permanent”: data is collected throughout the whole lifetime of the system, old information is overridden/corrected by newer input. • “Conceptual”: contains semantic relations between words and uses them to create concept definitions.

  9. Semantic memory Hierarchical model of semantic memory (Collins and Quillian, 1969), followed by most ontologies. Connectionist spreading activation model (Collins and Loftus, 1975), with mostly lateral connections. Our implementation is based on connectionist model, uses relational database and object access layer API. The database stores three types of data: • concepts, or objects being described; • keywords (features of concepts extracted from data sources); • relations between them. IS-A relation us used to build ontology tree, serving for activation spreading, i.e. features inheritance down the ontology tree. Types of relations (like “x IS y”, or “x CAN DO y” etc.) may be defined when input data is read from dictionaries and ontologies.

  10. Creating SM The API serves as a data access layer providing logical operations between raw data and higher application layers. Data stored in the database is mapped into application objects and the API allows for retrieving specific concepts/keywords. Two major types of data sources for semantic memory: • machine-readable structured dictionaries directly convertible into semantic memory data structures; • blocks of text, definitions of concepts from dictionaries/encyclopedias. 3 machine-readable data sources are used: • The Suggested Upper Merged Ontology (SUMO) and the the MId-Level Ontology (MILO), over 20,000 terms and 60,000 axioms. • WordNet lexicon, more than 200,000 words-sense pairs. • ConceptNet, concise knowledgebase with 200,000 assertions.

  11. Creating SM – free text WordNet hypernymic (a kind of … ) IS-A relation + Hyponym and meronym relations between synsets (converted into concept/concept relations), combined with ConceptNet relation such as: CapableOf, PropertyOf, PartOf, MadeOf ... Relations added only if in both Wordnet and Conceptnet. Free-text data: Merriam-Webster, WordNet and Tiscali. Whole word definitions are stored in SM linked to concepts. A set of most characteristic words from definitions of a given concept. For each concept definition, one set of words for each source dictionary is used, replaced with synset words, subset common to all 3 mapped back to synsets – these are most likely related to the initial concept. They were stored as a separate relation type. Articles and prepositions: removed using manually created stop-word list. Phrases were extracted using ApplePieParser + concept-phrase relations compared with concept-keyword, only phrases that matched keywords were used.

  12. Concept Description Vectors Drastic simplification: for some applications SM is used in a more efficient way using vector-based knowledge representation. Merging all types of relations => the most general one:“x IS RELATED TO y”, defining vector (semantic) space. {Concept, relations} => Concept Description Vector, CDV. Binary vector, shows which properties are related or have sense for a given concept (not the same as context vector). Semantic memory => CDV matrix, very sparse, easy storage of large amounts of semantic data. Search engines: {keywords} => concept descriptions (Web pages). CDV enable efficient implementation of reversed queries: find a unique subsets of properties for a given concept or a class of concepts = concept higher in ontology. What are the unique features of a sparrow? Proteoglycan? Neutrino?

  13. wCRK representation wCRK (weighted Concept - Relation Type - Keyword): Semantic memory theories are based on atomic logical elements: a typed relation (predicate) between concept (object) and keyword (feature) with weight value expressing the strength of this relation. wCRK forms a “word layer” of the system. Example: interactive exploration of the semantic space.

  14. Interactive semantic space

  15. Concept Description Vectors

  16. Talking Head SM is the brain, HIT needs a talking head and voice interface. Haptek’s PeoplePutty tools have been used (inexpensive) to create a 3-D talking head; only the simplest version is used. Haptek player is a plugin for Windows browsers, or embedded component in custom programs; both versions were used. High-fidelity natural voice synthesis with lips synchronization may be added to Haptek characters. Free MS Speech Engine, i.e. MS Speech API (SAPI 5) has been used to add text to speech synthesis and speech to text voice recognition. OGG prerecorded audio files may be played. Haptek movements, gestures, face expressions and animation sequences may be programmed and coordinated with speech using JavaScript, Visual Basic, Active-X Controls, C++, or ToolBook. Result: HIT that can interact with web pages, listen and talk, sending information both ways, hiding the text pages from the user. Interaction with Web pages is based on Microsoft .NET framework.

  17. HIT the Web Haptek avatar as a plug-in in WWW browser. Connect to web pages, read their contents, send queries and read answers from specific fields in web forms. Access Q/A pages, like MIT Start, or Brainboost that answer reasonably to many questions. “The HAL Nursery”, “the world's first Child-Machine Nursery”, Ai Research www.a-i.com, is hosting a collection of “Virtual Children”, or HAL personalities developed by many users through conversation. HAL is using reinforcement learning techniques to acquire language, through trial and error process similar to that infants are using. A child head with child voice makes it much more interesting to play with. Haptek heads may work with many chatterbots. There are many similar solutions, so we concentrated on the use of SM.

  18. Word games Word games that were popular before computer games took over. Word games are essential to the development of analytical thinking skills. Until recently computer technology was not sufficient to play such games. The 20 question game may be the next great challenge for AI, because it is more realistic than the unrestricted Turing test; a World Championship with human and software players in Singapore? How good are people in playing 20Q game? Performance of various models of semantic memory and episodic memory may be tested in this game in a realistic, difficult application. Asking questions to understand precisely what the user has in mind is critical for search engines and many other applications. Creating large-scale semantic memory with sufficient knowledge about all concepts is a great challenge.

  19. 20Q The goal of the 20 question game is to guess a concept that the opponent has in mind by asking appropriate questions. www.20q.net has a version that is now implemented in some toys! Based on concepts x question table T(C,Q) = usefulness of Q for C. Learns T(C,Q) values, increasing after successful games, decreasing after lost games. Guess: distance-based. SM does not assume fixed questions. Use of CDV admits only simplest form “Is it related to X?”, or “Can it be associated with X?”, where X = concept stored in the SM. Needs only to select a concept, not to build the whole question. Once the keyword has been selected it is possible to use the full power of semantic memory to analyze the type of relations and ask more sophisticated questions. How is the concept selected?

  20. Semantic Space exploration • Binary dictionary search for 20 questions with yes/no answers: 220 = 1 048 576 can be distinguished in principle, in practice search tree is never balanced. • Binary search – not acceptable in complex semantic applications • Semantic space can be search using context – based algorithm. Similar to a word game. • Concept space is narrowed by subsequent answers of the player. Information gained by using some keywords (features) is: where p(keyword = ni) is the fraction of concepts for which this keyword has value niSubspace of candidate concepts O(A) is selected by looking at the distance in the subspace of answers/keywords A used so far.

  21. Distance calculation x and y represent weights for CDV and ANS vectors and ||ANS|| is the Euclidean length of the answer vector. CDVn is a vector representing concept n and ANS is a partial vector of retrieved answers. O(A) is the subspace of candidate vectors, close to the ANS subspace We can deal with user mistakes including concepts with d > minimal

  22. Query Semantic memory Applications, eg. 20 questions game Humanized interface Store Part of speech tagger & phrase extractor verification On line dictionaries Parser Manual

  23. How to obtain semantic data? • CDV matrix for a single ontology reduced to animal kingdom was initially used to avoid storage size problems. • The first few steps find keywords with IG≈1. CDV vectors are too sparse, with 5-20, average 8, out of ~5000 keywords. In later stages IG is small, very few concepts eliminated. More information is needed in the semantic memory! • Wordnet is one of the best lexical sources • Relations for Semantic category: animal • 7543 objects and 1696 features • Truncated using word popularity rank: • IC – information content is an amount of appearances of the particular word in WordNet descriptions • GR - GoogleRank is the number of web pages returned by Google search engine for a given word • BNC are the words statistics taken from British National Norpus. - Semantic Space reduced to 889 objects and 420 features

  24. Active learning • Data from Wordnet are: • Not complete • Not common sense • Sometimes specialized concepts that a few people understand • Basic dialogs for obtaining new relations • I give up. Tell me what did you think of? • Tell me what is characteristic for <concept> ? • Knowledge correction: where: w0 – initial weight, initial knowledge ANS – answer given by user N – number of answers β- parameter scaling the importance of initial knowledge

  25. The game • Giraffe: [is vertebrate] Y,[is mammal] Y, [has hoof] Y, [is equine] N, [is bovine] N, [is deer] N, [is swine] N, [has horn] N, [has horn] N,[is sheep] N,[is antelope] N,[is bison] N. System correctly guess concept giraffe. • Yuppi i’ve won! Let’s talk about giraffe. Tell me what is characteristic forgiraffe? After entering keyword. Semantic memory is reorganised, and ready to play new games. • Lion: [is vertebrate] Y, [is mammal] Y, [has hoof] N, [has paw] Y, [is canine] N, [is cat] Y, [is wildcat] Y The different way for organizing concept lion in WordNet taxonomy, causes the game goes in wrong way and system fails guess this concept: [is leopard] N, [is painter] N, [is puma] N, [is lynx] N, [is lynx] N. I give up. What it was? Lion … After giving right answer system reorganizes its knowledge and next game for searching concept lion is finished with success: [is vertebrate] Y, [is mammal] Y, [has hoof] N, [has paw] Y, [is canine] N, [is cat] Y , [is wildcat] Y, [is leopard] N, [has mane] Y, I guess it is lion.

  26. Experimental results • How many games are needed to clarify semantic space? • Proportion of failed games Nf performed to achieve first success. • The semantic memory error: where Nsis amount of the games finished with success and N is total number of games; for searching first 10 concepts ~ 0.22 • How does it change during the learning process? Avg Density features / object

  27. A Salamander. If you do not know, ask Google!Quark page comes at the top … Puzzle generator Semantic memory may be used to invent automatically a large number of word puzzles that the avatar presents. This application selects a random concept from all concepts in the memory and searches for a minimal set of features necessary to uniquely define it; if many subsets are sufficient for unique definition one of them is selected randomly. It is an Amphibian, it is orange and has black spots. How do you call this animal? It has charm, it has spin, and it has charge. What is it?

  28. HIT – larger view … Learning Affective computing T-T-S synthesis Brain models Behavioralmodels Speech recognition HIT projects Cognitive Architectures Talking heads AI Robotics Cognitive science Graphics Lingu-bots A-Minds VR avatars Knowledgemodeling Info-retrieval WorkingMemory EpisodicMemory Semantic memory

  29. Future Language understanding is one of the key issues on the road to intelligent information management. Key problem identified: building good semantic memory! Many hierarchical ontologies exist, but no description of even simple objects. Wordnet definition of a cow: mature female of mammals of which the male is called ‘bull‘. Other dictionaries/info sources are not that useful too. One solution is an active search for correlations between possible properties and objects. Building good SM will enable many applications. Talking heads with artificial minds are the future of the Cyberspace! “A roadmap to human level intelligence” and many other initiatives discussing how to get there ...

More Related