SlideShare a Scribd company logo
Using Construction Grammar in  Conversational Systems Marie-Claire Jenkins, PhD Thesis (High level overview)
Overview This thesis was motivated by the machine's limitations in  understanding natural language and in forming responses. The limitations and complexities of current search engine querying was also a factor. Conversational systems are good for testing possible solutions and are useful on the web. We used methods that are not common in these systems: - Construction Grammar (CxG) - OWL ontologies - Lexical semantics - A new stemmer (Uea-Lite)
What I'm going to talk about ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Things I covered in my research: - Natural language understanding  - Natural language generation - Human computer interaction - Service oriented systems Things I didn't cover in my research: - Knowledge acquisition - Open domains - Affective behaviour - Everything else
Conversational systems They are more commonly referred to as "chatbots" or  “ Artificial Conversational Entities ” They converse with a user in natural language and simulate a human-human conversation. They need to: - "Understand ”  the user input - Retrieve relevant information - Generate a natural language response There are 3 different kinds of chatbots...
Social chatbots Their purpose is to chat freely about anything at all with a user, much like you would with a friend. They are used online for fun.
Educational chatbots Their purpose is to help the user learn about something such as a new language, history or geography. They are often used in schools
Service oriented chatbots Their purpose is to help customers find their way around the website and also to answer questions about their products & services.
How they work There are a variety of methods used but the most popular are: - Database driven - AIML (artificial intelligence markup language, xml based)  - Canned responses - Stochastic methods - Supervised learning - Named entity recognition - Templates
Phrase-based systems “ Phrase Based systems” are seen as  generalized templates  at the sentence level (like phrase structure rules) or at the discourse level. 1- Phrasal pattern selected [subject noun verb] 2 - Each part of the pattern is expanded [noun modifiers] 3 - When each phrasal pattern has been replaced by 1+ words –END They are very difficult to build because the phrasal interrelationships must be clearly specified otherwise there can be inappropriate phrase expansions.  
Feature-based systems In “Feature-based systems” each possible alternative is represented by a feature and each sentence is specified by them.   Sentence generation is achieved by using all of these features until the sentence is determined.   Features may include: positive/negative, past/present, statement/question… Strength:  any distinction in language can be a feature Weakness:  very hard to maintain feature inter-relationships and the control of feature selection.
Observations from live data Tests on dialogue from the  human-human  customer service system on a large commercial website reveal that there is no consistency in language or phrase formulation. There is a very small amount of Formulaic language (canned responses). A question was never formulated in the same way and never answered in the same way (apart from formulaicity). This makes it hard for us to produce templates or anticipate user utterances.
More Limitations Main issues with existing systems: - Scalability - Knowledge & information storage - User input disambiguation - Response generation (word order, vocabulary, etc...) - Knowledge/information retrieval - Anaphora - Managing the dialogue - Displaying appropriate behaviour (affective issues) - Knowledge assimilation - Evaluation
Turing test “ A machine is termed capable of thinking if it can, under certain prescribed conditions imitate a human by answering questions sufficiently well to deceive a human questioner for a reasonable period of time. ”  (Turing) Objections to the test include proving intelligence, "understanding" and other things. My personal opinion has changed since the beginning of my PhD research: “ The question of whether a computer can think is no more interesting than the question of whether a submarine can swim. ”  (Dijkstra)
Turing test illustration Wikipedia
XKCD
Loebner prize This  yearly contest  is run by Hugh Loebner who has offered a $100,000 prize for the 1st chatbot to pass the Turing test This test is controversial. Marvin Minsky  said : “ I do hope that someone will volunteer to violate this proscription so that Mr. Loebner will indeed revoke his stupid prize, save himself some money, and spare us the horror of this obnoxious and unproductive annual publicity campaign. ”
Loebner prize diagram Michael Mauldin- carnegie mellon  
John We built a conversational chatbot and entered it into the Loebner prize (2006). It was designed & built in 2 months and operated on a closed domain. Reason:  to run on a small database requiring little manual labour. We used ngrams, weighted responses, a vector approach, perl, Brill, UEA-Lite, wildcards, AIML We were a finalist and we learned that: - A small database worked for a small amount of time - A database system makes for laborious build and limited information (well used systems work much better) - Template methods are limited - Canned responses are awkward - AIML is restrictive
KIA: the HCI tests We designed a system made to research human-machine interaction and human behaviour: this is a test on humans and not the system We included functions that were meant to test user persistence with query repair, emotive response, language etc...  Results: users persist, are emotive, sensitive to interface design and more. Details available in our paper
KIA  –  a CxG & OWL driven system
Databases vs OWL ontologies: Databases focus on local semantics and ontologies on global semantics. In ontologies the semantics are explicit and in databases implicit.  Ontologies allow data to be reused whereas database schemas cannot be reused. Ontologies are portable between websites to facilitate maintenance and construction Restrictions in databases do not allow for all of the necessary relations to be built into the data.
Database ( Wordpress Bits ) Owl Ontology ( Richard Durban )
OWL flavour We used OWL (Web Ontology Language) as it is more expressive than other semantic web languages and is built to enable ontologies to be created easily.  It is a semantic markup language and an extension of RDF (Resource Description Framework).  There are different subsets of OWL: OWL Full, OWL Lite and OWL DL (Description Logic). We chose to use OWL DL.
Why Ontologies & why OWL DL? Taxonomies are also not as expansive as ontologies. “ At one extreme there are ontologies and the other mind maps and pathfinder networks, and in between taxonomies and browserable hierarchies ”. (Brewtser and Wilkes)  Ontologies have a greater potential for inference and a greater degree of formality.  OWL DL has stricter restrictions which are necessary in our type of system. It has maximum expressiveness without losing computational completeness (all entailments are will be computed) and decidability (all computations will finish in finite time) of reasoning systems.
OWL Ontology example: Koala
What do we store in there? - All of the domain knowledge (e.g all about Koalas) - The collection of constructions (commonly used when discussing koalas) - Canned responses (formulaic language)
KIA system domain knowledge
Construction Grammar  It is a cognitive linguistic method and it is: - Constraint based - Generative - Non-derivational - A monostratal grammatical model - Incorporates the cognitive and interactional foundations of language - Consists of taxonomies of families of constructions - Uses entire constructions as the primary unit of grammar - Is a pairing of form and meaning (metonomic) - Frames used in CxG != regular frames because the argument structure types invoke frames which designate event types - The verb alone is not the main unit of meaning, the construction itself is
Constructions Words Sentences Constructions make sense in computing
Example of CxG Semantics:  relational predicate involving a singer   Syntactics:  predicate requires arguments and ``Heather'' is the subject Generative Grammar Construction Grammar
Advantages of CxG - Adapts to changing language patterns easily - Takes into consideration both semantics and syntactics - Constructions are easier to manage than words as the atomic unit - Allows for integration into bigger collections of constructions - Can be computed
UEA-Lite stemmer After testing the system with all available stemmers, we realised that we needed to design our own to facilitate topic/construction detection. UEA-Lite stems conservatively to orthographically correct word forms and recognizes words which do not need to be stemmed. There is a  Perl, Java  and  Ruby  version More information  here (an updated paper to follow soon)
Machine learning  ,[object Object],[object Object],[object Object],[object Object]
Algorithms -  Jaccard Distance  to weight the constructions (how often different constructions are found in conjunction, partial or complete) -  Naive Bayes algorithm  clusters all of the constructions according to their different features in our training set (requires little training data) Once the data has been processed through the Naive Bayes algorithm we know which constructions are often found with others, and in what order. We not only look at the syntax but also at the semantic aspect both in isolation and in conjunction with each other. The role of the classifier is to determine which categories future constructions belong to, and also to tell us which constructions are a likely match to a query.
Naïve Bayes for CxG P (Constructions) doesn't change over time. Naive Bayes estimates a multinomial distribution over categories, which is the prior distribution of categories We can therefore say that: Best category [ArgaMax cat in cats] = P (constructions ¦ cat) (P (cat))  If c1, c2, ... cn are the constructions in the document, then: Best category [ArgaMax cat in cats] = P(c1|cat)*P(c2|cat)*...*P(cn|cat)*P(cat)
System diagram There are many more  components to the system than presented in this  presentation as you can see.
Evaluation methods There are not any robust evaluation methods for conversational systems but we found that a mixture of the following worked well: - Human evaluation (feedback form) - " Pourpre ”  to evaluate sentence complexity (Jimmy Lin) - Expected vs Given response score Evaluation is not finished as yet but the initial results are encouraging with good knowledge retrieval and construction selection.
Things that didn't work ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Fluid construction Grammar (FCG) (also didn't work!) - Bi-directional (using rules) - Selects meanings and maps them into the real world. - "fluid" because it takes into consideration the fact that users change and update their grammars often.  - User input can be broken down syntactically in order to gain meaning from the grammatical components, whilst also being able to map the semantic relationships BUT : not developed enough to work well in our system  Also:  bi-directional rules are very hard to write
Some Outcomes & Learnings - Construction Grammar is a useful method for NLU & NLG - OWL ontologies are well suited to these systems - Stemming affects the system greatly - Fluid CxG is not practical at this time - Better evaluation methods need to be developed - Turing test is not useful as it does not prove machine intelligence or understanding - User perception is a primordial area of research
Applications & Future work - Assisted search - Summarization systems - Content creation - Speech systems - Sentiment analysis - More powerful AI module - Anaphora resolution - Open domain testing - Improved machine learning - Further work on query disambiguation methods
Thank you Find me at: http://www.scienceforseo.com http://twitter.com/missmcj Google reader

More Related Content

What's hot

Structural & Transformational Grammars
Structural & Transformational GrammarsStructural & Transformational Grammars
Structural & Transformational Grammars
trinorei22
 
Cognitive Linguistics: The Case Of Find
Cognitive Linguistics: The Case Of FindCognitive Linguistics: The Case Of Find
Cognitive Linguistics: The Case Of Find
JESSIE GRACE RUBRICO
 
Modular Ontologies - A Formal Investigation of Semantics and Expressivity
Modular Ontologies - A Formal Investigation of Semantics and ExpressivityModular Ontologies - A Formal Investigation of Semantics and Expressivity
Modular Ontologies - A Formal Investigation of Semantics and ExpressivityJie Bao
 
Nlp (1)
Nlp (1)Nlp (1)
An introduction to compositional models in distributional semantics
An introduction to compositional models in distributional semanticsAn introduction to compositional models in distributional semantics
An introduction to compositional models in distributional semanticsAndre Freitas
 
A COMPUTATIONAL APPROACH FOR ANALYZING INTER-SENTENTIAL ANAPHORIC PRONOUNS IN...
A COMPUTATIONAL APPROACH FOR ANALYZING INTER-SENTENTIAL ANAPHORIC PRONOUNS IN...A COMPUTATIONAL APPROACH FOR ANALYZING INTER-SENTENTIAL ANAPHORIC PRONOUNS IN...
A COMPUTATIONAL APPROACH FOR ANALYZING INTER-SENTENTIAL ANAPHORIC PRONOUNS IN...
ijnlc
 
Flexible Concept Mapping
Flexible Concept MappingFlexible Concept Mapping
Flexible Concept Mapping
Alfredo Tifi
 
Ic analysis by shadab akhter
Ic analysis by shadab akhter Ic analysis by shadab akhter
Ic analysis by shadab akhter
. .
 
Introduction to Distributional Semantics
Introduction to Distributional SemanticsIntroduction to Distributional Semantics
Introduction to Distributional SemanticsAndre Freitas
 
The Cognitive Perspective in Teaching Chinese measure words
The Cognitive Perspective in Teaching Chinese measure words The Cognitive Perspective in Teaching Chinese measure words
The Cognitive Perspective in Teaching Chinese measure words
UAB - Universitat Autònoma de Barcelona
 
Minimalist program
Minimalist programMinimalist program
Minimalist program
Amusan Kayode
 
I C ANALYSIS
I C ANALYSISI C ANALYSIS
I C ANALYSIS
Shranti Hake
 
FCA-MERGE: Bottom-Up Merging of Ontologies
FCA-MERGE: Bottom-Up Merging of OntologiesFCA-MERGE: Bottom-Up Merging of Ontologies
FCA-MERGE: Bottom-Up Merging of Ontologiesalemarrena
 
Clause complex (maira, sofia, mercedes)
Clause complex (maira, sofia, mercedes)Clause complex (maira, sofia, mercedes)
Clause complex (maira, sofia, mercedes)rominacheme
 
Distributional semantics
Distributional semanticsDistributional semantics
Distributional semantics
Rabindra Nath Nandi
 
Cognitive linguistics
Cognitive linguisticsCognitive linguistics
Cognitive linguistics
Adel Thamery
 
Minimalist program
Minimalist programMinimalist program
Minimalist program
RabbiaAzam
 
CLUE-Aligner: An Alignment Tool to Annotate Pairs of Paraphrastic and Transla...
CLUE-Aligner: An Alignment Tool to Annotate Pairs of Paraphrastic and Transla...CLUE-Aligner: An Alignment Tool to Annotate Pairs of Paraphrastic and Transla...
CLUE-Aligner: An Alignment Tool to Annotate Pairs of Paraphrastic and Transla...
INESC-ID (Spoken Language Systems Laboratory - L2F)
 
Lecture5 Meaning
Lecture5 MeaningLecture5 Meaning
Lecture5 Meaning
Adel Thamery
 

What's hot (20)

Structural & Transformational Grammars
Structural & Transformational GrammarsStructural & Transformational Grammars
Structural & Transformational Grammars
 
Cognitive Linguistics: The Case Of Find
Cognitive Linguistics: The Case Of FindCognitive Linguistics: The Case Of Find
Cognitive Linguistics: The Case Of Find
 
Modular Ontologies - A Formal Investigation of Semantics and Expressivity
Modular Ontologies - A Formal Investigation of Semantics and ExpressivityModular Ontologies - A Formal Investigation of Semantics and Expressivity
Modular Ontologies - A Formal Investigation of Semantics and Expressivity
 
Nlp (1)
Nlp (1)Nlp (1)
Nlp (1)
 
An introduction to compositional models in distributional semantics
An introduction to compositional models in distributional semanticsAn introduction to compositional models in distributional semantics
An introduction to compositional models in distributional semantics
 
A COMPUTATIONAL APPROACH FOR ANALYZING INTER-SENTENTIAL ANAPHORIC PRONOUNS IN...
A COMPUTATIONAL APPROACH FOR ANALYZING INTER-SENTENTIAL ANAPHORIC PRONOUNS IN...A COMPUTATIONAL APPROACH FOR ANALYZING INTER-SENTENTIAL ANAPHORIC PRONOUNS IN...
A COMPUTATIONAL APPROACH FOR ANALYZING INTER-SENTENTIAL ANAPHORIC PRONOUNS IN...
 
Flexible Concept Mapping
Flexible Concept MappingFlexible Concept Mapping
Flexible Concept Mapping
 
Ic analysis by shadab akhter
Ic analysis by shadab akhter Ic analysis by shadab akhter
Ic analysis by shadab akhter
 
Introduction to Distributional Semantics
Introduction to Distributional SemanticsIntroduction to Distributional Semantics
Introduction to Distributional Semantics
 
The Cognitive Perspective in Teaching Chinese measure words
The Cognitive Perspective in Teaching Chinese measure words The Cognitive Perspective in Teaching Chinese measure words
The Cognitive Perspective in Teaching Chinese measure words
 
Minimalist program
Minimalist programMinimalist program
Minimalist program
 
I C ANALYSIS
I C ANALYSISI C ANALYSIS
I C ANALYSIS
 
FCA-MERGE: Bottom-Up Merging of Ontologies
FCA-MERGE: Bottom-Up Merging of OntologiesFCA-MERGE: Bottom-Up Merging of Ontologies
FCA-MERGE: Bottom-Up Merging of Ontologies
 
Clause complex (maira, sofia, mercedes)
Clause complex (maira, sofia, mercedes)Clause complex (maira, sofia, mercedes)
Clause complex (maira, sofia, mercedes)
 
Distributional semantics
Distributional semanticsDistributional semantics
Distributional semantics
 
Cognitive linguistics
Cognitive linguisticsCognitive linguistics
Cognitive linguistics
 
Minimalist program
Minimalist programMinimalist program
Minimalist program
 
Incrementality
IncrementalityIncrementality
Incrementality
 
CLUE-Aligner: An Alignment Tool to Annotate Pairs of Paraphrastic and Transla...
CLUE-Aligner: An Alignment Tool to Annotate Pairs of Paraphrastic and Transla...CLUE-Aligner: An Alignment Tool to Annotate Pairs of Paraphrastic and Transla...
CLUE-Aligner: An Alignment Tool to Annotate Pairs of Paraphrastic and Transla...
 
Lecture5 Meaning
Lecture5 MeaningLecture5 Meaning
Lecture5 Meaning
 

Viewers also liked

Charles fillmore s cases
Charles fillmore s casesCharles fillmore s cases
Charles fillmore s casesajosrdzr
 
Language acquisition2
Language acquisition2Language acquisition2
Language acquisition2
ilugones
 
04051000126
0405100012604051000126
04051000126Lyly Le
 
Protocolo azahalia santanapool&sandychiblanco
Protocolo azahalia santanapool&sandychiblancoProtocolo azahalia santanapool&sandychiblanco
Protocolo azahalia santanapool&sandychiblancoSandy CB
 
Powerpoint of Graduation Thesis of English Major
Powerpoint of Graduation Thesis of  English MajorPowerpoint of Graduation Thesis of  English Major
Powerpoint of Graduation Thesis of English Major
Phi Pham
 
Gaby and vanesa thesis
Gaby and vanesa thesisGaby and vanesa thesis
Gaby and vanesa thesis
Diana Ashandy Pool Antonio
 
101 Conversational User Interfaces
101 Conversational User Interfaces 101 Conversational User Interfaces
101 Conversational User Interfaces
Redweb Ltd
 
The Age of Conversational Agents
The Age of Conversational AgentsThe Age of Conversational Agents
The Age of Conversational Agents
Faction XYZ
 
Conversational Collaboration: How Messaging, App Integration, and Chatbots ar...
Conversational Collaboration: How Messaging, App Integration, and Chatbots ar...Conversational Collaboration: How Messaging, App Integration, and Chatbots ar...
Conversational Collaboration: How Messaging, App Integration, and Chatbots ar...
Dion Hinchcliffe
 
Graduation thesis of English Major
Graduation thesis of English MajorGraduation thesis of English Major
Graduation thesis of English Major
Phi Pham
 
Conversational interfaces - beyond the hype
Conversational interfaces - beyond the hypeConversational interfaces - beyond the hype
Conversational interfaces - beyond the hype
Matthieu Varagnat
 
AI and Python: Developing a Conversational Interface using Python
AI and Python: Developing a Conversational Interface using PythonAI and Python: Developing a Conversational Interface using Python
AI and Python: Developing a Conversational Interface using Python
amyiris
 
Chatbots, Conversational Interfaces, and the Rise of Messaging platforms
Chatbots, Conversational Interfaces, and the Rise of Messaging platformsChatbots, Conversational Interfaces, and the Rise of Messaging platforms
Chatbots, Conversational Interfaces, and the Rise of Messaging platforms
Sean Johnson
 
Research report traditional grammar vs functional grammar and teaching of gr...
Research report traditional grammar vs functional grammar and teaching of  gr...Research report traditional grammar vs functional grammar and teaching of  gr...
Research report traditional grammar vs functional grammar and teaching of gr...Rai Shoaib Ali
 
factors affecting boarding,non-boarding and transient boarding education stud...
factors affecting boarding,non-boarding and transient boarding education stud...factors affecting boarding,non-boarding and transient boarding education stud...
factors affecting boarding,non-boarding and transient boarding education stud...
rica mahinay
 
Yunit 3 istruktura ng wika
Yunit 3  istruktura ng wikaYunit 3  istruktura ng wika
Yunit 3 istruktura ng wika
Rita Mae Odrada
 
Strategic Intervention Material (SIM) Filipino-NOLI ME TANGERE
Strategic Intervention Material (SIM) Filipino-NOLI ME TANGEREStrategic Intervention Material (SIM) Filipino-NOLI ME TANGERE
Strategic Intervention Material (SIM) Filipino-NOLI ME TANGERE
Sophia Marie Verdeflor
 

Viewers also liked (20)

Charles fillmore s cases
Charles fillmore s casesCharles fillmore s cases
Charles fillmore s cases
 
Language acquisition2
Language acquisition2Language acquisition2
Language acquisition2
 
04051000126
0405100012604051000126
04051000126
 
Protocolo azahalia santanapool&sandychiblanco
Protocolo azahalia santanapool&sandychiblancoProtocolo azahalia santanapool&sandychiblanco
Protocolo azahalia santanapool&sandychiblanco
 
Grammar ; abridgment
Grammar ; abridgmentGrammar ; abridgment
Grammar ; abridgment
 
Powerpoint of Graduation Thesis of English Major
Powerpoint of Graduation Thesis of  English MajorPowerpoint of Graduation Thesis of  English Major
Powerpoint of Graduation Thesis of English Major
 
Gaby and vanesa thesis
Gaby and vanesa thesisGaby and vanesa thesis
Gaby and vanesa thesis
 
Tesis sergio rejon
Tesis sergio rejonTesis sergio rejon
Tesis sergio rejon
 
101 Conversational User Interfaces
101 Conversational User Interfaces 101 Conversational User Interfaces
101 Conversational User Interfaces
 
The Age of Conversational Agents
The Age of Conversational AgentsThe Age of Conversational Agents
The Age of Conversational Agents
 
Conversational Collaboration: How Messaging, App Integration, and Chatbots ar...
Conversational Collaboration: How Messaging, App Integration, and Chatbots ar...Conversational Collaboration: How Messaging, App Integration, and Chatbots ar...
Conversational Collaboration: How Messaging, App Integration, and Chatbots ar...
 
Graduation thesis of English Major
Graduation thesis of English MajorGraduation thesis of English Major
Graduation thesis of English Major
 
Conversational interfaces - beyond the hype
Conversational interfaces - beyond the hypeConversational interfaces - beyond the hype
Conversational interfaces - beyond the hype
 
AI and Python: Developing a Conversational Interface using Python
AI and Python: Developing a Conversational Interface using PythonAI and Python: Developing a Conversational Interface using Python
AI and Python: Developing a Conversational Interface using Python
 
Protocolo gabriela buenfil
Protocolo gabriela buenfilProtocolo gabriela buenfil
Protocolo gabriela buenfil
 
Chatbots, Conversational Interfaces, and the Rise of Messaging platforms
Chatbots, Conversational Interfaces, and the Rise of Messaging platformsChatbots, Conversational Interfaces, and the Rise of Messaging platforms
Chatbots, Conversational Interfaces, and the Rise of Messaging platforms
 
Research report traditional grammar vs functional grammar and teaching of gr...
Research report traditional grammar vs functional grammar and teaching of  gr...Research report traditional grammar vs functional grammar and teaching of  gr...
Research report traditional grammar vs functional grammar and teaching of gr...
 
factors affecting boarding,non-boarding and transient boarding education stud...
factors affecting boarding,non-boarding and transient boarding education stud...factors affecting boarding,non-boarding and transient boarding education stud...
factors affecting boarding,non-boarding and transient boarding education stud...
 
Yunit 3 istruktura ng wika
Yunit 3  istruktura ng wikaYunit 3  istruktura ng wika
Yunit 3 istruktura ng wika
 
Strategic Intervention Material (SIM) Filipino-NOLI ME TANGERE
Strategic Intervention Material (SIM) Filipino-NOLI ME TANGEREStrategic Intervention Material (SIM) Filipino-NOLI ME TANGERE
Strategic Intervention Material (SIM) Filipino-NOLI ME TANGERE
 

Similar to Using construction grammar in conversational systems

NLP_A Chat-Bot_answering_queries_of_UT-Dallas_Students
NLP_A Chat-Bot_answering_queries_of_UT-Dallas_StudentsNLP_A Chat-Bot_answering_queries_of_UT-Dallas_Students
NLP_A Chat-Bot_answering_queries_of_UT-Dallas_StudentsHimanshu kandwal
 
ISWC2023-McGuinnessTWC16x9FinalShort.pdf
ISWC2023-McGuinnessTWC16x9FinalShort.pdfISWC2023-McGuinnessTWC16x9FinalShort.pdf
ISWC2023-McGuinnessTWC16x9FinalShort.pdf
Deborah McGuinness
 
Big Data and Natural Language Processing
Big Data and Natural Language ProcessingBig Data and Natural Language Processing
Big Data and Natural Language Processing
Michel Bruley
 
Frame-Script and Predicate logic.pptx
Frame-Script and Predicate logic.pptxFrame-Script and Predicate logic.pptx
Frame-Script and Predicate logic.pptx
nilesh405711
 
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUECOMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
Journal For Research
 
[IJET-V1I6P17] Authors : Mrs.R.Kalpana, Mrs.P.Padmapriya
[IJET-V1I6P17] Authors : Mrs.R.Kalpana, Mrs.P.Padmapriya[IJET-V1I6P17] Authors : Mrs.R.Kalpana, Mrs.P.Padmapriya
[IJET-V1I6P17] Authors : Mrs.R.Kalpana, Mrs.P.Padmapriya
IJET - International Journal of Engineering and Techniques
 
INTELLIGENT QUERY PROCESSING IN MALAYALAM
INTELLIGENT QUERY PROCESSING IN MALAYALAMINTELLIGENT QUERY PROCESSING IN MALAYALAM
INTELLIGENT QUERY PROCESSING IN MALAYALAM
ijcsa
 
IRJET- Querying Database using Natural Language Interface
IRJET-  	  Querying Database using Natural Language InterfaceIRJET-  	  Querying Database using Natural Language Interface
IRJET- Querying Database using Natural Language Interface
IRJET Journal
 
NE7012- SOCIAL NETWORK ANALYSIS
NE7012- SOCIAL NETWORK ANALYSISNE7012- SOCIAL NETWORK ANALYSIS
NE7012- SOCIAL NETWORK ANALYSIS
rathnaarul
 
ESWC SS 2012 - Tuesday Tutorial Elena Simperl: Creating and Using Ontologies
ESWC SS 2012 - Tuesday Tutorial Elena Simperl: Creating and Using OntologiesESWC SS 2012 - Tuesday Tutorial Elena Simperl: Creating and Using Ontologies
ESWC SS 2012 - Tuesday Tutorial Elena Simperl: Creating and Using Ontologieseswcsummerschool
 
A neural probabilistic language model
A neural probabilistic language modelA neural probabilistic language model
A neural probabilistic language model
c sharada
 
NLP Tasks and Applications.ppt useful in
NLP Tasks and Applications.ppt useful inNLP Tasks and Applications.ppt useful in
NLP Tasks and Applications.ppt useful in
Kumari Naveen
 
lect36-tasks.ppt
lect36-tasks.pptlect36-tasks.ppt
lect36-tasks.ppt
HaHa501620
 
Complexity
ComplexityComplexity
Complexity
Rajesh Kumar
 
Breaking down the AI magic of ChatGPT: A technologist's lens to its powerful ...
Breaking down the AI magic of ChatGPT: A technologist's lens to its powerful ...Breaking down the AI magic of ChatGPT: A technologist's lens to its powerful ...
Breaking down the AI magic of ChatGPT: A technologist's lens to its powerful ...
rahul_net
 
Natural Language Processing (NLP)
Natural Language Processing (NLP)Natural Language Processing (NLP)
Natural Language Processing (NLP)
Yuriy Guts
 
Multilayered paper prototyping for user concept modeling
Multilayered paper prototyping for user concept modelingMultilayered paper prototyping for user concept modeling
Multilayered paper prototyping for user concept modeling
UKOLN (dev), University of Bath
 
Association Rule Mining Based Extraction of Semantic Relations Using Markov ...
Association Rule Mining Based Extraction of  Semantic Relations Using Markov ...Association Rule Mining Based Extraction of  Semantic Relations Using Markov ...
Association Rule Mining Based Extraction of Semantic Relations Using Markov ...
dannyijwest
 

Similar to Using construction grammar in conversational systems (20)

NLP_A Chat-Bot_answering_queries_of_UT-Dallas_Students
NLP_A Chat-Bot_answering_queries_of_UT-Dallas_StudentsNLP_A Chat-Bot_answering_queries_of_UT-Dallas_Students
NLP_A Chat-Bot_answering_queries_of_UT-Dallas_Students
 
ISWC2023-McGuinnessTWC16x9FinalShort.pdf
ISWC2023-McGuinnessTWC16x9FinalShort.pdfISWC2023-McGuinnessTWC16x9FinalShort.pdf
ISWC2023-McGuinnessTWC16x9FinalShort.pdf
 
Big Data and Natural Language Processing
Big Data and Natural Language ProcessingBig Data and Natural Language Processing
Big Data and Natural Language Processing
 
Frame-Script and Predicate logic.pptx
Frame-Script and Predicate logic.pptxFrame-Script and Predicate logic.pptx
Frame-Script and Predicate logic.pptx
 
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUECOMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
COMPREHENSIVE ANALYSIS OF NATURAL LANGUAGE PROCESSING TECHNIQUE
 
The impact of standardized terminologies and domain-ontologies in multilingua...
The impact of standardized terminologies and domain-ontologies in multilingua...The impact of standardized terminologies and domain-ontologies in multilingua...
The impact of standardized terminologies and domain-ontologies in multilingua...
 
[IJET-V1I6P17] Authors : Mrs.R.Kalpana, Mrs.P.Padmapriya
[IJET-V1I6P17] Authors : Mrs.R.Kalpana, Mrs.P.Padmapriya[IJET-V1I6P17] Authors : Mrs.R.Kalpana, Mrs.P.Padmapriya
[IJET-V1I6P17] Authors : Mrs.R.Kalpana, Mrs.P.Padmapriya
 
INTELLIGENT QUERY PROCESSING IN MALAYALAM
INTELLIGENT QUERY PROCESSING IN MALAYALAMINTELLIGENT QUERY PROCESSING IN MALAYALAM
INTELLIGENT QUERY PROCESSING IN MALAYALAM
 
FinalReport
FinalReportFinalReport
FinalReport
 
IRJET- Querying Database using Natural Language Interface
IRJET-  	  Querying Database using Natural Language InterfaceIRJET-  	  Querying Database using Natural Language Interface
IRJET- Querying Database using Natural Language Interface
 
NE7012- SOCIAL NETWORK ANALYSIS
NE7012- SOCIAL NETWORK ANALYSISNE7012- SOCIAL NETWORK ANALYSIS
NE7012- SOCIAL NETWORK ANALYSIS
 
ESWC SS 2012 - Tuesday Tutorial Elena Simperl: Creating and Using Ontologies
ESWC SS 2012 - Tuesday Tutorial Elena Simperl: Creating and Using OntologiesESWC SS 2012 - Tuesday Tutorial Elena Simperl: Creating and Using Ontologies
ESWC SS 2012 - Tuesday Tutorial Elena Simperl: Creating and Using Ontologies
 
A neural probabilistic language model
A neural probabilistic language modelA neural probabilistic language model
A neural probabilistic language model
 
NLP Tasks and Applications.ppt useful in
NLP Tasks and Applications.ppt useful inNLP Tasks and Applications.ppt useful in
NLP Tasks and Applications.ppt useful in
 
lect36-tasks.ppt
lect36-tasks.pptlect36-tasks.ppt
lect36-tasks.ppt
 
Complexity
ComplexityComplexity
Complexity
 
Breaking down the AI magic of ChatGPT: A technologist's lens to its powerful ...
Breaking down the AI magic of ChatGPT: A technologist's lens to its powerful ...Breaking down the AI magic of ChatGPT: A technologist's lens to its powerful ...
Breaking down the AI magic of ChatGPT: A technologist's lens to its powerful ...
 
Natural Language Processing (NLP)
Natural Language Processing (NLP)Natural Language Processing (NLP)
Natural Language Processing (NLP)
 
Multilayered paper prototyping for user concept modeling
Multilayered paper prototyping for user concept modelingMultilayered paper prototyping for user concept modeling
Multilayered paper prototyping for user concept modeling
 
Association Rule Mining Based Extraction of Semantic Relations Using Markov ...
Association Rule Mining Based Extraction of  Semantic Relations Using Markov ...Association Rule Mining Based Extraction of  Semantic Relations Using Markov ...
Association Rule Mining Based Extraction of Semantic Relations Using Markov ...
 

More from CJ Jenkins

I am an experience designer
I am an experience designer I am an experience designer
I am an experience designer CJ Jenkins
 
How Sentiment Analysis works
How Sentiment Analysis worksHow Sentiment Analysis works
How Sentiment Analysis works
CJ Jenkins
 
Knowledgebase vs Database
Knowledgebase vs DatabaseKnowledgebase vs Database
Knowledgebase vs Database
CJ Jenkins
 
Building a semantic website
Building a semantic websiteBuilding a semantic website
Building a semantic website
CJ Jenkins
 
Search Engine Spiders
Search Engine SpidersSearch Engine Spiders
Search Engine Spiders
CJ Jenkins
 
Twitter for business
Twitter for businessTwitter for business
Twitter for business
CJ Jenkins
 
The search engine index
The search engine indexThe search engine index
The search engine index
CJ Jenkins
 

More from CJ Jenkins (7)

I am an experience designer
I am an experience designer I am an experience designer
I am an experience designer
 
How Sentiment Analysis works
How Sentiment Analysis worksHow Sentiment Analysis works
How Sentiment Analysis works
 
Knowledgebase vs Database
Knowledgebase vs DatabaseKnowledgebase vs Database
Knowledgebase vs Database
 
Building a semantic website
Building a semantic websiteBuilding a semantic website
Building a semantic website
 
Search Engine Spiders
Search Engine SpidersSearch Engine Spiders
Search Engine Spiders
 
Twitter for business
Twitter for businessTwitter for business
Twitter for business
 
The search engine index
The search engine indexThe search engine index
The search engine index
 

Recently uploaded

Free Complete Python - A step towards Data Science
Free Complete Python - A step towards Data ScienceFree Complete Python - A step towards Data Science
Free Complete Python - A step towards Data Science
RinaMondal9
 
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
Neo4j
 
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
James Anderson
 
GridMate - End to end testing is a critical piece to ensure quality and avoid...
GridMate - End to end testing is a critical piece to ensure quality and avoid...GridMate - End to end testing is a critical piece to ensure quality and avoid...
GridMate - End to end testing is a critical piece to ensure quality and avoid...
ThomasParaiso2
 
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdfFIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance
 
Elevating Tactical DDD Patterns Through Object Calisthenics
Elevating Tactical DDD Patterns Through Object CalisthenicsElevating Tactical DDD Patterns Through Object Calisthenics
Elevating Tactical DDD Patterns Through Object Calisthenics
Dorra BARTAGUIZ
 
GraphSummit Singapore | The Art of the Possible with Graph - Q2 2024
GraphSummit Singapore | The Art of the  Possible with Graph - Q2 2024GraphSummit Singapore | The Art of the  Possible with Graph - Q2 2024
GraphSummit Singapore | The Art of the Possible with Graph - Q2 2024
Neo4j
 
Introduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - CybersecurityIntroduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - Cybersecurity
mikeeftimakis1
 
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
Neo4j
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
KatiaHIMEUR1
 
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdfSAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
Peter Spielvogel
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Aggregage
 
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdfFIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance
 
State of ICS and IoT Cyber Threat Landscape Report 2024 preview
State of ICS and IoT Cyber Threat Landscape Report 2024 previewState of ICS and IoT Cyber Threat Landscape Report 2024 preview
State of ICS and IoT Cyber Threat Landscape Report 2024 preview
Prayukth K V
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
sonjaschweigert1
 
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
DanBrown980551
 
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdfSmart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
91mobiles
 
National Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practicesNational Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practices
Quotidiano Piemontese
 
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
SOFTTECHHUB
 

Recently uploaded (20)

Free Complete Python - A step towards Data Science
Free Complete Python - A step towards Data ScienceFree Complete Python - A step towards Data Science
Free Complete Python - A step towards Data Science
 
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
 
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
 
GridMate - End to end testing is a critical piece to ensure quality and avoid...
GridMate - End to end testing is a critical piece to ensure quality and avoid...GridMate - End to end testing is a critical piece to ensure quality and avoid...
GridMate - End to end testing is a critical piece to ensure quality and avoid...
 
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdfFIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
 
Elevating Tactical DDD Patterns Through Object Calisthenics
Elevating Tactical DDD Patterns Through Object CalisthenicsElevating Tactical DDD Patterns Through Object Calisthenics
Elevating Tactical DDD Patterns Through Object Calisthenics
 
GraphSummit Singapore | The Art of the Possible with Graph - Q2 2024
GraphSummit Singapore | The Art of the  Possible with Graph - Q2 2024GraphSummit Singapore | The Art of the  Possible with Graph - Q2 2024
GraphSummit Singapore | The Art of the Possible with Graph - Q2 2024
 
Introduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - CybersecurityIntroduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - Cybersecurity
 
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
 
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdfSAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
 
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdfFIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
 
State of ICS and IoT Cyber Threat Landscape Report 2024 preview
State of ICS and IoT Cyber Threat Landscape Report 2024 previewState of ICS and IoT Cyber Threat Landscape Report 2024 preview
State of ICS and IoT Cyber Threat Landscape Report 2024 preview
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
 
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
 
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdfSmart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
 
National Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practicesNational Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practices
 
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
 

Using construction grammar in conversational systems

  • 1. Using Construction Grammar in Conversational Systems Marie-Claire Jenkins, PhD Thesis (High level overview)
  • 2. Overview This thesis was motivated by the machine's limitations in  understanding natural language and in forming responses. The limitations and complexities of current search engine querying was also a factor. Conversational systems are good for testing possible solutions and are useful on the web. We used methods that are not common in these systems: - Construction Grammar (CxG) - OWL ontologies - Lexical semantics - A new stemmer (Uea-Lite)
  • 3.
  • 4. Things I covered in my research: - Natural language understanding - Natural language generation - Human computer interaction - Service oriented systems Things I didn't cover in my research: - Knowledge acquisition - Open domains - Affective behaviour - Everything else
  • 5. Conversational systems They are more commonly referred to as "chatbots" or “ Artificial Conversational Entities ” They converse with a user in natural language and simulate a human-human conversation. They need to: - "Understand ” the user input - Retrieve relevant information - Generate a natural language response There are 3 different kinds of chatbots...
  • 6. Social chatbots Their purpose is to chat freely about anything at all with a user, much like you would with a friend. They are used online for fun.
  • 7. Educational chatbots Their purpose is to help the user learn about something such as a new language, history or geography. They are often used in schools
  • 8. Service oriented chatbots Their purpose is to help customers find their way around the website and also to answer questions about their products & services.
  • 9. How they work There are a variety of methods used but the most popular are: - Database driven - AIML (artificial intelligence markup language, xml based) - Canned responses - Stochastic methods - Supervised learning - Named entity recognition - Templates
  • 10. Phrase-based systems “ Phrase Based systems” are seen as generalized templates at the sentence level (like phrase structure rules) or at the discourse level. 1- Phrasal pattern selected [subject noun verb] 2 - Each part of the pattern is expanded [noun modifiers] 3 - When each phrasal pattern has been replaced by 1+ words –END They are very difficult to build because the phrasal interrelationships must be clearly specified otherwise there can be inappropriate phrase expansions.  
  • 11. Feature-based systems In “Feature-based systems” each possible alternative is represented by a feature and each sentence is specified by them.   Sentence generation is achieved by using all of these features until the sentence is determined.   Features may include: positive/negative, past/present, statement/question… Strength: any distinction in language can be a feature Weakness: very hard to maintain feature inter-relationships and the control of feature selection.
  • 12. Observations from live data Tests on dialogue from the human-human customer service system on a large commercial website reveal that there is no consistency in language or phrase formulation. There is a very small amount of Formulaic language (canned responses). A question was never formulated in the same way and never answered in the same way (apart from formulaicity). This makes it hard for us to produce templates or anticipate user utterances.
  • 13. More Limitations Main issues with existing systems: - Scalability - Knowledge & information storage - User input disambiguation - Response generation (word order, vocabulary, etc...) - Knowledge/information retrieval - Anaphora - Managing the dialogue - Displaying appropriate behaviour (affective issues) - Knowledge assimilation - Evaluation
  • 14. Turing test “ A machine is termed capable of thinking if it can, under certain prescribed conditions imitate a human by answering questions sufficiently well to deceive a human questioner for a reasonable period of time. ” (Turing) Objections to the test include proving intelligence, "understanding" and other things. My personal opinion has changed since the beginning of my PhD research: “ The question of whether a computer can think is no more interesting than the question of whether a submarine can swim. ” (Dijkstra)
  • 16. XKCD
  • 17. Loebner prize This yearly contest is run by Hugh Loebner who has offered a $100,000 prize for the 1st chatbot to pass the Turing test This test is controversial. Marvin Minsky said : “ I do hope that someone will volunteer to violate this proscription so that Mr. Loebner will indeed revoke his stupid prize, save himself some money, and spare us the horror of this obnoxious and unproductive annual publicity campaign. ”
  • 18. Loebner prize diagram Michael Mauldin- carnegie mellon  
  • 19. John We built a conversational chatbot and entered it into the Loebner prize (2006). It was designed & built in 2 months and operated on a closed domain. Reason: to run on a small database requiring little manual labour. We used ngrams, weighted responses, a vector approach, perl, Brill, UEA-Lite, wildcards, AIML We were a finalist and we learned that: - A small database worked for a small amount of time - A database system makes for laborious build and limited information (well used systems work much better) - Template methods are limited - Canned responses are awkward - AIML is restrictive
  • 20. KIA: the HCI tests We designed a system made to research human-machine interaction and human behaviour: this is a test on humans and not the system We included functions that were meant to test user persistence with query repair, emotive response, language etc... Results: users persist, are emotive, sensitive to interface design and more. Details available in our paper
  • 21. KIA – a CxG & OWL driven system
  • 22. Databases vs OWL ontologies: Databases focus on local semantics and ontologies on global semantics. In ontologies the semantics are explicit and in databases implicit. Ontologies allow data to be reused whereas database schemas cannot be reused. Ontologies are portable between websites to facilitate maintenance and construction Restrictions in databases do not allow for all of the necessary relations to be built into the data.
  • 23. Database ( Wordpress Bits ) Owl Ontology ( Richard Durban )
  • 24. OWL flavour We used OWL (Web Ontology Language) as it is more expressive than other semantic web languages and is built to enable ontologies to be created easily. It is a semantic markup language and an extension of RDF (Resource Description Framework). There are different subsets of OWL: OWL Full, OWL Lite and OWL DL (Description Logic). We chose to use OWL DL.
  • 25. Why Ontologies & why OWL DL? Taxonomies are also not as expansive as ontologies. “ At one extreme there are ontologies and the other mind maps and pathfinder networks, and in between taxonomies and browserable hierarchies ”. (Brewtser and Wilkes) Ontologies have a greater potential for inference and a greater degree of formality. OWL DL has stricter restrictions which are necessary in our type of system. It has maximum expressiveness without losing computational completeness (all entailments are will be computed) and decidability (all computations will finish in finite time) of reasoning systems.
  • 27. What do we store in there? - All of the domain knowledge (e.g all about Koalas) - The collection of constructions (commonly used when discussing koalas) - Canned responses (formulaic language)
  • 28. KIA system domain knowledge
  • 29. Construction Grammar  It is a cognitive linguistic method and it is: - Constraint based - Generative - Non-derivational - A monostratal grammatical model - Incorporates the cognitive and interactional foundations of language - Consists of taxonomies of families of constructions - Uses entire constructions as the primary unit of grammar - Is a pairing of form and meaning (metonomic) - Frames used in CxG != regular frames because the argument structure types invoke frames which designate event types - The verb alone is not the main unit of meaning, the construction itself is
  • 30. Constructions Words Sentences Constructions make sense in computing
  • 31. Example of CxG Semantics: relational predicate involving a singer   Syntactics: predicate requires arguments and ``Heather'' is the subject Generative Grammar Construction Grammar
  • 32. Advantages of CxG - Adapts to changing language patterns easily - Takes into consideration both semantics and syntactics - Constructions are easier to manage than words as the atomic unit - Allows for integration into bigger collections of constructions - Can be computed
  • 33. UEA-Lite stemmer After testing the system with all available stemmers, we realised that we needed to design our own to facilitate topic/construction detection. UEA-Lite stems conservatively to orthographically correct word forms and recognizes words which do not need to be stemmed. There is a Perl, Java and Ruby version More information here (an updated paper to follow soon)
  • 34.
  • 35. Algorithms - Jaccard Distance to weight the constructions (how often different constructions are found in conjunction, partial or complete) - Naive Bayes algorithm clusters all of the constructions according to their different features in our training set (requires little training data) Once the data has been processed through the Naive Bayes algorithm we know which constructions are often found with others, and in what order. We not only look at the syntax but also at the semantic aspect both in isolation and in conjunction with each other. The role of the classifier is to determine which categories future constructions belong to, and also to tell us which constructions are a likely match to a query.
  • 36. Naïve Bayes for CxG P (Constructions) doesn't change over time. Naive Bayes estimates a multinomial distribution over categories, which is the prior distribution of categories We can therefore say that: Best category [ArgaMax cat in cats] = P (constructions ¦ cat) (P (cat))  If c1, c2, ... cn are the constructions in the document, then: Best category [ArgaMax cat in cats] = P(c1|cat)*P(c2|cat)*...*P(cn|cat)*P(cat)
  • 37. System diagram There are many more components to the system than presented in this presentation as you can see.
  • 38. Evaluation methods There are not any robust evaluation methods for conversational systems but we found that a mixture of the following worked well: - Human evaluation (feedback form) - " Pourpre ” to evaluate sentence complexity (Jimmy Lin) - Expected vs Given response score Evaluation is not finished as yet but the initial results are encouraging with good knowledge retrieval and construction selection.
  • 39.
  • 40. Fluid construction Grammar (FCG) (also didn't work!) - Bi-directional (using rules) - Selects meanings and maps them into the real world. - "fluid" because it takes into consideration the fact that users change and update their grammars often. - User input can be broken down syntactically in order to gain meaning from the grammatical components, whilst also being able to map the semantic relationships BUT : not developed enough to work well in our system  Also: bi-directional rules are very hard to write
  • 41. Some Outcomes & Learnings - Construction Grammar is a useful method for NLU & NLG - OWL ontologies are well suited to these systems - Stemming affects the system greatly - Fluid CxG is not practical at this time - Better evaluation methods need to be developed - Turing test is not useful as it does not prove machine intelligence or understanding - User perception is a primordial area of research
  • 42. Applications & Future work - Assisted search - Summarization systems - Content creation - Speech systems - Sentiment analysis - More powerful AI module - Anaphora resolution - Open domain testing - Improved machine learning - Further work on query disambiguation methods
  • 43. Thank you Find me at: http://www.scienceforseo.com http://twitter.com/missmcj Google reader