Junhwi Choi

Modeling confirmations for example-based dialog management

This paper proposes a method to model confirmations for example-based dialog management. To enable the system to provide a confirmation to the user in an appropriate time, we employed a multiple dialog state representation approach for keeping track of user input uncertainty and implemented a confirmation agent which decides when the information gathered from the user contains an error. We developed a car navigation dialog system to evaluate our proposed method. Evaluations with simulated dialogs show our approach is useful for handling misunderstanding errors in example-based dialog management.

Acquisition and Use of Long-Term Memory for Personalized Dialog Systems

This study introduces a personalization framework for dialog systems. Our system automatically collects user-related facts (i.e. triples) from user input sentences and stores the facts in one-shot memory. The system also keeps track of changes in user interests. Extracted triples and entities (i.e. NP-chunks) are stored in a personal knowledge base (PKB) and a forgetting model manages their retention (i.e. interest). System responses can be modified by applying user-related facts to the one-shot memory. A relevance score of a system response is proposed to select responses that include high-retention triples and entities, or frequently used responses. We used Movie-Dic corpus to construct a simple dialog system and train PKBs. The retention sum of responses was increased by adopting the PKB, and the number of inappropriate responses was decreased by adopting relevance score. The system gave some personalized responses, while maintaining its performance (i.e. appropriateness of responses).

Seamless error correction interface for voice word processor

In this paper, we propose an error correction interface for a voice word processor. This correction interface includes user intention understanding and automatic error region detection. For accurate correction, we include a confirmation process that includes an error region control command and a re-uttering command. We evaluate the performance of the user intention understanding first, and we evaluate the effectiveness of our interface compare to a general two-step error correction interface.

A Two-Step Approach for Efficient Domain Selection in Multi-Domain Dialog Systems

This paper discusses a domain selection method for multi-domain dialog systems to generate the most appropriate system utterance in response to a user utterance. We present a two-step approach for efficient domain selection. In our proposed approach, the domain candidates are listed in descending order of scores and then each domain is verified by content-based filtering. When we applied our method, the accuracy increased and the time cost decreased compared to baseline methods.

Conversational Knowledge Teaching Agent that uses a Knowledge Base

When implementing a conversational educational teaching agent, user-intent understanding and dialog management in a dialog system are not sufficient to give users educational information. In this paper, we propose a conversational educational teaching agent that gives users some educational information or triggers interests on educational contents. The proposed system not only converses with a user but also answer questions that the user asked or asks some educational questions by integrating a dialog system with a knowledge base. We used the Wikipedia corpus to learn the weights between two entities and embedding of properties to calculate similarities for the selection of system questions and answers.

Neural sentence embedding using only in-domain sentences for out-of-domain sentence detection in dialog systems

Only in-domain sentences are used to train out-of-domain sentence detection.A long short-term memory network is used to extract features from sentences.Domain-category analysis is used as an auxiliary task.A two-channel approach is applied to word representations.An autoencoder is used for one-class classification. To ensure satisfactory user experience, dialog systems must be able to determine whether an input sentence is in-domain (ID) or out-of-domain (OOD). We assume that only ID sentences are available as training data because collecting enough OOD sentences in an unbiased way is a laborious and time-consuming job. This paper proposes a novel neural sentence embedding method that represents sentences in a low-dimensional continuous vector space that emphasizes aspects that distinguish ID cases from OOD cases. We first used a large set of unlabeled text to pre-train word representations that are used to initialize neural sentence embedding. Then we used domain-category analysis as an auxiliary task to train neural sentence embedding for OOD sentence detection. After the sentence representations were learned, we used them to train an autoencoder aimed at OOD sentence detection. We evaluated our method by experimentally comparing it to the state-of-the-art methods in an eight-domain dialog system; our proposed method achieved the highest accuracy in all tests.

ASR Error Management Using RNN Based Syllable Prediction for Spoken Dialog Applications

We proposed automatic speech recognition (ASR) error management method using recurrent neural network (RNN) based syllable prediction for spoken dialog applications. ASR errors are detected and corrected by syllable prediction. For accurate prediction of a next syllable, we used a current syllable, previous syllable context, and phonetic information of next syllable which is given by ASR error. The proposed method can correct ASR errors only with a text corpus which is used for training of the target application, and it means that the method is independent to the ASR engine. The method is general and can be applied to any speech based application such as spoken dialog systems.

Engine-Independent ASR Error Management for Dialog Systems

This paper describes a method of ASR (automatic speech recognition) engine independent error correction for a dialog system. The proposed method can correct ASR errors only with a text corpus which is used for training of the target dialog system, and it means that the method is independent of the ASR engine. We evaluated our method on two test corpora (Korean and English) that are parallel corpora including ASR results and their correct transcriptions. Overall results indicate that the method decreases the word error rate of the ASR results and recovers the errors in the important attributes of the dialog system. The method is general and can also be applied to the other speech based applications such as voice question-answering and speech information extraction systems.

ASR Independent Hybrid Recurrent Neural Network Based Error Correction for Dialog System Applications

We proposed an automatic speech recognition (ASR) error correction method using hybrid word sequence matching and recurrent neural network for dialog system applications. Basically, the ASR errors are corrected by the word sequence matching whereas the remaining OOV (out of vocabulary) errors are corrected by the secondary method which uses a recurrent neural network based syllable prediction. We evaluated our method on a test parallel corpus (Korean) including ASR results and their correct transcriptions. Overall result indicates that the method effectively decreases the word error rate of the ASR results. The proposed method can correct ASR errors only with a text corpus without their speech recognition results, which means that the method is independent to the ASR engine. The method is general and can be applied to any speech based application such as spoken dialog systems.

Hierarchical Dirichlet Process Topic Modeling for Large Number of Answer Types Classification in Open domain Question Answering

We propose a new method that uses the Hierarchical Dirichlet Process (HDP) to classify a large number of answer types for a question posed using natural language. We used the HDP model to build a classifier that assigns test questions to certain clusters, then computes similarity among the questions within the same cluster. Our answer-type classifier finds the n-best similar training questions to the test questions and classifies the test question’s answer type as the majority of the n-best training question’s answer type. The proposed method achieved similar accuracy and lower sensitivity to the presence of a large number of answer types than existing methods that use classification algorithms with same features. Also, we can guarantee that appropriate answer type can be among the ranked answer types with high recall.