113,877 Hits in 4.3 sec

Exploiting distance based similarity in topic models for user intent detection

Asli Celikyilmaz, Dilek Hakkani-Tur, Gokhan Tur, Ashley Fidler, Dustin Hillard
2011 2011 IEEE Workshop on Automatic Speech Recognition & Understanding  
One of the main components of spoken language understanding is intent detection, which allows user goals to be identified.  ...  We demonstrate that our model outperforms standard topic models based on bag-of-words assumption.  ...  Without an intent bearing n-gram determiner, the intent detection task becomes intrinsically challenging, not only because there are no a priori constraints on what the user might say, but also because  ... 
doi:10.1109/asru.2011.6163969 dblp:conf/asru/CelikyilmazHTFH11 fatcat:z4hyheebrfajbiwyyzpnfup5lm


2009 International Journal of Uncertainty Fuzziness and Knowledge-Based Systems  
This approach is a further step on building more streamline user-adaptive newswire search, based on topic and trend detection systems [9], able to deal in a userfriendly way with the permanently changing  ...  Detecting and understanding user behavior and relating user intentions to emerging topic trends in news data streams still continues to remain a huge challenge for making search engines in "real-time"  ...  Our first goal is to detect the user intentions from the user queries logs.  ... 
doi:10.1142/s0218488509006030 fatcat:lkvl4eg45fezrbrmw7tbmqiscq

Information Navigation System Based on POMDP that Tracks User Focus

Koichiro Yoshino, Tatsuya Kawahara
2014 Proceedings of the 15th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL)  
The POMDP selects one of six action classes; each action class is implemented with its own module.  ...  Jason Williams for his valuable and detailed advice to improve this paper on SIGDIAL mentoring program. This work was supported by Grant-in-Aid for JSPS Fellows 25-4537.  ...  User focus detection based on CRF To detect the user focus, we use a conditional random field (CRF) 1 .  ... 
doi:10.3115/v1/w14-4305 dblp:conf/sigdial/YoshinoK14 fatcat:phpexdtwpzbu3pulkfset4bqu4

Review of Intent Detection Methods in the Human-Machine Dialogue System

Jiao Liu, Yanling Li, Min Lin
2019 Journal of Physics, Conference Series  
detection task, so as to promote the research of multi-intent detection methods based on deep neural network.  ...  This paper mainly analyzes, compares and summarizes the deep learning methods applied in the research of intent detection in recent years, and further considers how to apply deep learning model to multi-intent  ...  Intent detection based on the combination of deep learning models.  ... 
doi:10.1088/1742-6596/1267/1/012059 fatcat:bgbnmku5gnafbimg2i3aanuma4

Alquist 2.0: Alexa Prize Socialbot Based on Sub-Dialogue Models [article]

Jan Pichl, Petr Marek, Jakub Konrád, Martin Matulík, Jan Šedivý
2020 arXiv   pre-print
The sub-dialogues can be triggered according to the topic hierarchy or a user intent which allows the bot to create a unique experience during each session.  ...  We introduce a system leveraging ontology-based topic structure called topic nodes.  ...  Note that a new sub-dialogue is triggered based on the entity and intent only if the topic switch is detected (based on the contextual data) as described in subsection 3.4.  ... 
arXiv:2011.03259v1 fatcat:gq2mccx4n5dozeuo2uv426hy54

The essence of time

Stewart Whiting
2012 Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '12  
Key challenges to be addressed in this work include: a suitable source and method for event detection and tracking, an intent-aware ranking approach and an evaluation methodology.  ...  As users become exposed to this information, time plays a central role in prompting both information authorship and seeking activities.  ...  We propose an evaluation methodology based on query-log mining and crowdsourcing for on-going relevance assessment.  ... 
doi:10.1145/2348283.2348424 dblp:conf/sigir/Whiting12 fatcat:gd7ae6d7czb3hldzw6zzi3mpdy

Knowledge Discovery from Big Data for Intrusion Detection Using LDA

Jingwei Huang, Zbigniew Kalbarczyk, David M. Nicol
2014 2014 IEEE International Congress on Big Data  
We identify the "hidden" patterns of operations conducted by both normal users and malicious users from a large volume of network/systems logs, by mapping this problem to the topic modeling problem and  ...  This new approach potentially completes the strength of signature-based and anomaly-based methods.  ...  What a user is trying to do is unknown to the security monitoring system; however, although a single event does not reveal the user's intent, a collection of logged events may reveal the intent.  ... 
doi:10.1109/bigdata.congress.2014.111 dblp:conf/bigdata/HuangKN14 fatcat:iviasghrpbhgtkundvmt3her64

An algorithm for analyzing personalized online commercial intention

Derek Hao Hu, Qiang Yang, Ying Li
2008 Proceedings of the 2nd International Workshop on Data Mining and Audience Intelligence for Advertising - ADKDD '08  
In this paper, we present a novel algorithm, which we name as POINT, for the Personalized Onlinecommercial INTention detection based on a skip-chain conditional random field model.  ...  The only existing research work, as far as we know, has been done to automatically detect online commercial intention purely based on the issued queries, without considering the Web user's information.  ...  In [11] , user goals, or user interests are estimated from the user's clickthrough histories, based on the Topic-Sensitive PageRank scores of these pages.  ... 
doi:10.1145/1517472.1517476 fatcat:5chzdzfg35bjtjs3ccyuk2egry

News Navigation System Based on Proactive Dialogue Strategy [chapter]

Koichiro Yoshino, Tatsuya Kawahara
2015 Natural Language Dialog Systems and Intelligent Assistants  
An experimental result shows that the proposed system based on POMDP and user focus tracking can interact with users effectively by selecting the most appropriate dialogue modules.  ...  It is composed of several modules that interact with users in different manners. Both the system and the user can take an initiative of dialogue depending on the specification of the user interest.  ...  The dialogue manager selects a module (ac-tion decision) based on an input of a user intention.  ... 
doi:10.1007/978-3-319-19291-8_2 fatcat:k3olzjau7bh5zjschrl2o7xm7a

RubyStar: A Non-Task-Oriented Mixture Model Dialog System [article]

Huiting Liu, Tao Lin, Hanfei Sun, Weijian Lin, Chih-Wei Chang, Teng Zhong, Alexander Rudnicky
2017 arXiv   pre-print
RubyStar conducts a non-task-oriented conversation on general topics by using an ensemble of rule-based, retrieval-based and generative methods.  ...  Topic detection, engagement monitoring, and context tracking are used for managing interaction.  ...  Topic Detection To guide the conversation on a more comfortable and engaging course, topic detection is used to track context and topics over time.  ... 
arXiv:1711.02781v3 fatcat:orqlnnx46jamxngd2akuoun56y

Domain-Specific Modeling of User Knowledge in Informational Search Sessions

Rui Tang, Ran Yu, Markus Rokicki, Ralph Ewerth, Stefan Dietze
2021 International Conference on Information and Knowledge Management  
Users frequently search on the Web to fulfill information needs with learning intent. In this context, usefulness of the search results depends strongly on the knowledge state of the user.  ...  However, the impact of knowledge domains of the search topics on a user's learning process have not been adequately explored.  ...  Informational search sessions involve an inherent learning intent, i.e. the desire of a user to acquire knowledge or information with respect to a particular topic, assumed to be present on one or more  ... 
dblp:conf/cikm/Tang0RED21 fatcat:cktlki73enbh7oz25nkwjocakq

Topic-centric Classification of Twitter User's Political Orientation

Anjie Fang, Iadh Ounis, Philip Habel, Craig Macdonald, Nut Limsopatham
2015 Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR '15  
Our new classifier identifies the closest topic for each unseen tweet, based on those topics identified in the training data.  ...  We show that the views communicated on these topics can inform us of the individuals' voting intentions ("Yes" vs. "No").  ...  We propose a technique to analyse the voting intentions of users, based on data mining and machine learning approaches.  ... 
doi:10.1145/2766462.2767833 dblp:conf/sigir/FangOHML15 fatcat:asov5pxurfgk3fn743dp5xg7um

Alexa Prize — State of the Art in Conversational AI

Chandra Khatri, Anu Venkatesh, Behnam Hedayatnia, Raefer Gabriel, Ashwin Ram, Rohit Prasad
2018 The AI Magazine  
popular topics for 20 minutes.  ...  This article reports on the research conducted over the 2017-2018 year.  ...  Intent Detection Intents represent the goal of a user for a given utterance, and the dialogue system needs to detect it to act and respond appropriately to that utterance.  ... 
doi:10.1609/aimag.v39i3.2810 fatcat:kuprycf7gbdvjiuf33b5ylmfli

Behavior-driven clustering of queries into topics

Luca Maria Aiello, Debora Donato, Umut Ozertem, Filippo Menczer
2011 Proceedings of the 20th ACM international conference on Information and knowledge management - CIKM '11  
Our method extracts topics from the query log by merging missions, i.e., activity fragments that express a coherent user intent, on the basis of their topical affinity.  ...  It can also summarize topically-coherent missions that occur far away from each other, thus enabling a more compact user profiling on a topical basis.  ...  user intent.  ... 
doi:10.1145/2063576.2063775 dblp:conf/cikm/AielloDOM11 fatcat:i6scd6grwfctvp4tqq4alhebmm

Active Subtopic Detection in Multitopic Data

Benjamin Bergner, Georg Krempl
2016 International Conference on Knowledge Management and Knowledge Technologies  
Each approach is evaluated based on the number of subtopics that are found in the same predefined, closed topic (countries).  ...  For the task of detecting subtopics with user guidance, clustering by intent (CBI) has recently been proposed. However, this approach is limited to single-topic environments.  ...  Acknowledgements We would like to thank Andreas Nürnberger, Daniel Kottke, and George Forman for insightful discussions on this topic.  ... 
dblp:conf/iknow/BergnerK16 fatcat:xfjbebwfazhrvmwl3ptjoqnpxi
« Previous Showing results 1 — 15 out of 113,877 results