Ontology Induction (Chen et al., 2013 & 2014) Frame-semantic parsing on ASR results (Das et al., 2013) • frame slot candidate • lexical unit slot filler 1st Issue: How to induce domain- specific concepts? locale_by_u se food expensivene ss seekin g relational_quan tity PREP_FOR PREP_FOR NN AMOD AMOD AMOD desirin g DOBJ typ e food pricerang e DOBJ AMOD AMOD AMOD tas k area PREP_IN •Domain: restaurant recommendation in an in-car setting (WER = 37%) o Dialogue slots: addr, area, food, phone, postcode, pricerange, task, type speak on topic add r are a foo d phon e part orientational direction locale part inner outer food origin contactin g postco de pric e rang e tas k typ e sendin g commerce scenario expensiveness range seekin g desiri ng locati ng locale by use building reference ontology w/ the most frequent dependencie s Can a dialogue system automatically learn open domain knowledge and then understand users? Unsupervised Learning and Modeling of Knowledge and Intent for Spoken Dialogue Systems Yun-Nung (Vivian) Chen [email protected] Goal Framework Restaurant Asking Conversati ons targe t food price seekin g quantity PREP_FOR PREP_FOR NN AMOD AMOD AMOD Domain-Specific Ontology Unlabeled Collection Knowledge Acquisition Ontology Induction Structure Learning Domain-Specific Ontology price=“cheap”, target=“restaurant” behavior=navigation SLU Modeling by MF SLU Component “can i have a cheap restaurant” Semantic Decoding Behavior Prediction Knowledge Acquisition SLU Modeling by Matrix Factorization Structure Learning (Chen et al., 2015a) Typed syntactic dependencies on ASR https://github.com/yvchen/MRRW/ ccomp amod dobj nsubj det cani havea chea p restaura nt capabilit y expensivene ss locale_by_ use induced ontolog y Semantic Decoding (Chen et al., 2015b) • concept semantic slot Behavior Prediction • concept user behavior 1) Given unlabeled conversations, how can a system automatically induce and organize domain-specific concepts? 2) With the automatically acquired knowledge, how can a system understand utterances? 1 Utterance 1 i would like a cheap restaurant Feature Observation Semantic Concept (Slot / Behavior) T r a i n … … … cheap restaurant food expensiveness 1 target 1 1 find a restaurant for chinese food Utterance 2 1 1 food 1 1 1 T e s t 1 1 .97 .90 .95 .85 .93 .92 .98 .05 .05 Feature Relation Model Concept Relation Model Reasoning with Matrix Factorization Semantic Concept Induction SLU Model Semantic Representation “can I have a cheap restaurant” Ontology Induction Unlabele d Collecti on SLU Modeling by MF F f F s Feature Model R f R s Relation Propagation Model Feature Relation Model Concept Relation Model . Knowledge Acquisition Structu re Learnin g o Relation Propagation Model Feature Knowledge Graph Concept Knowledge Graph Assumption: The domain-specific features/concepts have more dependency to each other. Relation matrices allow each node to propagate scores to its neighbors in the knowledge graph, so that domain-specific features/concepts have higher scores during training. 2nd Issue: Hidden semantics cannot be observed but may benefit understanding performance. Good! Good! ? Feature Relation Model Concept Relation Model R f R s ‧ 1 Feature Concept T r a i n cheap restaurant food expensiven ess 1 locale_by_ use 1 1 1 1 food 1 1 1 T e s t 1 1 Ontology Induction i like 1 1 capabil ity 1 Utterance 1 i would like a cheap restaurant … … find a restaurant with chinese food Utteranc e 2 show me a list of cheap restaurants Test Utterance .9 0 .9 7 .9 5 .8 5 hidden semanti cs o Matrix Factorization (MF) Model implicit feedback Objective: 1 + ¿¿ − − MF learns a set of well-ranked concepts per utterance. Chen et al., “Unsupervised Induction and Filling of Semantic Slots for Spoken Dialogue Systems Using Frame-Semantic Parsing,” in Proc. of ASRU, 2013. Chen et al., “Leveraging Frame Semantics and Distributional Semantics for Unsupervised Semantic Slot Induction in Spoken Dialogue Systems,” in Proc. of SLT, 2014. Chen et al., “Jointly Modeling Inter-Slot Relations by Random Walk on Knowledge Graphs for Unsupervised Spoken Language Understanding,” in Proc. of NAACL, 2015a. Chen et al., “Matrix Factorization with Knowledge Graph Propagation for Unsupervised Spoken Language Understanding,” in Proc. of ACL-IJCNLP, 2015b.