From Text To Reasoning - Marko Grobelnik - SWANK Workshop Stanford - 16 Apr 2014
Download as PPTX, PDF1 like909 views
The document discusses the goal of natural language processing moving beyond "processing" text to truly understanding text through the use of world models. It provides an example of a top-down world model called Cyc and how such a model can be used to disambiguate text and allow for reasoning beyond the information directly stated. Micro-reading is discussed as understanding individual documents through extracting logical assertions and allowing for inference, with examples provided. Finally, some of the key challenges for artificial intelligence in developing comprehensive world models and achieving true understanding are outlined.
From Text To Reasoning - Marko Grobelnik - SWANK Workshop Stanford - 16 Apr 2014
- 1. From Text to Reasoning Marko Grobelnik Jozef Stefan Institute / Cycorp Europe, Slovenia SWANK Workshop, Stanford, Apr 16th 2014Thanks to Michael Witbrock, Janez Starc, Luka Bradesko, Blaz Fortuna
- 2. Reflection on what should be the goal of NLP • The (mostly) forgotten long term aim of NLP is to understand the text • …and not so much ‘processing’ itself (as NLP suggests) • The curse of shallow solutions working well enough for too many problems, made people (and researchers) happy for too long • …as much as information retrieval and text mining are useful, they delayed development of “text understanding”
- 3. Language vs. World • …if we agree with the above statement, then at this point in time, we have ‘language’, but the ‘world’ is more or less missing • So – so what a ‘world’ or ‘world model’ could be?
- 4. CYC KNOWLEDGE BASE Thing Universe isa isa Celestial Body isa located in Planet subclass Earth isa Animal isa Human subclas s Physics Money Mathematics Chemistry Time Learning FoodVehicles Event Education School Language LoveEmotions Going for a walk Death Cat Euro Working Words Driving RainStabbing someone Nature Tree Hatred Fear Physics Time Learning Vehicles Event Education School Emotions Going for a walk Death Cat EuroWords Driving Rain Stabbing someone Nature Tree Hatred Fear Planet Earth isa Human Physics Money Mathematics Chemistry Time Learning FoodVehicles Event Education Languag e LoveEmotions Going for a walk Cat Euro Working Words Driving Rain Tree Hatred Fear Learning Vehicles Event Education School Emotions Euro Driving Stabbing someone Hatred Fear Creating a World Model (top-down approach - Cyc)
- 5. Model of the world… • …beyond surface knowledge • …to interconnect contextualized fragments Why? • To make reasoning capable of connecting isolated fragments of knowledge • To derive new knowledge beyond materialized factual knowledge World model Top-down KA Bottom-up KA Multimodal data Why we need a World model?
- 6. Disambiguation with a world model (CycKB)World model used as a set of common-sense semantic constraints to disambiguate text
- 7. One of the challenges for the future: Micro-reading • It is “easier” to understand millions of documents than one document • …reading and understanding a single document is micro-reading • The following experiment is on how much knowledge we can extract from individual documents • …extraction is in a form of first order inferentially productive Cyc logic • …allowing us full reasoning to identify new facts • …minimizing human involvement, optimizing precision and recall Document Assertions Reasoning Dialogue
- 8. Example of text and extracted Cyc assertions (1/2) Automatically Extracted Assertions: • (isa ?V1 ProsecutingEvent) • (agent ?V1 RudyGiuliani) • (genls Entity Agent) • (isa RudyGiuliani Agent) • (isa RudyGiuliani Entity) • (isa ?V3 OrganizingEvent) • (patient ?V3 (IntersectionFn OrganizedCrime WallStreet)) • (isa (IntersectionFn OrganizedCrime WallStreet) Patient) • (genls Entity Patient) • (isa OrganizedCrime Patient) • (isa OrganizedCrime Entity) • (isa WallStreet Patient) • (isa WallStreet Entity) Sentence: He prosecuted a number of high-profile cases, including ones against organized crime and Wall_Street financiers.
- 9. Example of text and extracted Cyc assertions (2/2) Automatically Extracted Assertions: • (isa ?V1 SubstitutingEvent) • (temporal ?V1 Lincoln) • (genls Entity Agent) • (isa Lincoln Agent) • (genls Person Entity) • (isa Lincoln Entity) • (isa Lincoln Person) • (isa ?V3 SucceedingEvent) • (temporal ?V3 Grant) • (isa Grant Agent) • (isa Grant Entity) • (isa Grant Person) Sentence: Each time a general failed, Lincoln substituted another until finally Grant succeeded in 1865.
- 10. Reasoning on extracted assertions (Cyc) Query: (and (isa ?Per Person) (birthDate ?Per ?BD) (occursBefore ?BD WorldWarII) (thereExistsAtLeast 2 ?Role (lifeRole ?Per ?Role) (roleInIndustry ?Role FilmIndustry) ) ) Answers: Sir Derek_George_Jacobi Sir Alexander_Korda Victor Lonzo_Fleming John_Francis_Junkin Cornel_Wilde George_Stevens Bertrand_Blier NL Query: People born before World War II who had at least two roles in the film industry KB?
- 11. Knowledge Capture Knowledge Use Rule: (implies (and (isa ?VENUE FoodTruck-Organization) (lastVenue ?USER ?VENUE) (suggestionsForCuriousCatQuestionType FoodTruckSecondaryTypeOfPlace- CuriousCatQuestion ?SUGGESTIONLIST)) (curiousCatWantsToAskUser ?USER (secondaryTypeOfPlace ?VENUE FoodTruck-Organization ?TYPE) ?SUGGESTIONLIST)) Witbrock, M., Bradeško, L., 2013, Conversational Computation in Michelucci, Pietro (Ed.) Handbook of Human Computation, 531-543. Intelligent SIRI: http://curiouscat.cc/
- 12. Some of the AI challenges for next years • Background knowledge in a form of a World Model • …to have knowledge contextualized • Representing and scalable reasoning knowledge with operational soft logic • …to decrease brittleness of logic and increase scale • Economically viable structured knowledge acquisition with high precision and recall • …to increase the reach of what we can acquire • Emphasizing understanding vs. applying black box models