125 Hits in 3.3 sec

When Topic Models Disagree

Lucas Sterckx, Thomas Demeester, Johannes Deleu, Chris Develder
2015 Proceedings of the 24th International Conference on World Wide Web - WWW '15 Companion  
We show that averaging multiple topic models, inferred from different corpora, leads to more accurate keyphrases than when using a single topic model and other state-of-the-art techniques.  ...  We explore how the unsupervised extraction of topic-related keywords benefits from combining multiple topic models.  ...  CONCLUSION In this paper we showed ongoing work demonstrating the benefit of combining multiple topic models for Automatic Keyphrase Extraction.  ... 
doi:10.1145/2740908.2742731 dblp:conf/www/SterckxDDD15a fatcat:hxg6rlxtsjfyvjdvy44ivk7qay

A Review of Keyphrase Extraction [article]

Eirini Papagiannopoulou, Grigorios Tsoumakas
2019 arXiv   pre-print
Keyphrase extraction is a textual information processing task concerned with the automatic extraction of representative and characteristic phrases from a document that express all the key aspects of its  ...  This article introduces keyphrase extraction, provides a well-structured review of the existing work, offers interesting insights on the different evaluation approaches, highlights open issues and presents  ...  Particularly, the models disagree when they are trained on different corpora, as there is a difference in contexts between the corpora.  ... 
arXiv:1905.05044v2 fatcat:xeweqtrjrfbefi2h5g42uld4pe

TAN-NTM: Topic Attention Networks for Neural Topic Modeling [article]

Madhur Panwar, Shashank Shailabh, Milan Aggarwal, Balaji Krishnamurthy
2021 arXiv   pre-print
Further, we show that our method learns better latent document-topic features compared to existing topic models through improvement on two downstream tasks: document classification and topic guided keyphrase  ...  Topic models have been widely used to learn text representations and gain insight into document corpora.  ...  Word intrusion measures the presence of those words (called intruder words) which disagree with the semantics of the topic.  ... 
arXiv:2012.01524v2 fatcat:qzupfp6zzvcejdhqhswx7whniu

Topic Segmentation and Labeling in Asynchronous Conversations

S. Joty, G. Carenini, R. T. Ng
2013 The Journal of Artificial Intelligence Research  
For topic segmentation, we propose two novel unsupervised models that exploit the fine-grained conversational structure, and a novel graph-theoretic supervised model that combines lexical, conversational  ...  Empirical evaluation shows that the segmentation and the labeling performed by our best models beat the state-of-the-art, and are highly correlated with human annotations.  ...  We also compare our model with two state-of-the-art keyphrase extraction methods.  ... 
doi:10.1613/jair.3940 fatcat:s264hyayknfpjiaqorgo67f32e

Text Classification Model Explainability for Keyword Extraction – Towards Keyword-Based Summarization of Nursing Care Episodes [chapter]

Akseli Reunamo, Laura-Maria Peltonen, Reetta Mustonen, Minttu Saari, Tapio Salakoski, Sanna Salanterä, Hans Moen
2022 Studies in Health Technology and Informatics  
This study aims to extract keywords and phrases that provide an intuitive overview of the content in multiple nursing entries in EHRs written during individual patients' care episodes.  ...  The proposed keyword extraction method is used to generate keyword summaries from 40 patients' care episodes and its performance is compared to a baseline method based on word embeddings combined with  ...  Next, using the classification model we apply a XAI method to extract the most predictive words for the topic heading with the highest confidence.  ... 
doi:10.3233/shti220154 pmid:35673093 fatcat:jualx7rq45gyrdq6ubucylfmya

Identifying important concepts from medical documents

Quanzhi Li, Yi-Fang Brook Wu
2006 Journal of Biomedical Informatics  
KIP combines two functions: noun phrase extraction and keyphrase identification. The former automatically extracts noun phrases from medical literature as keyphrase candidates.  ...  In this paper, we present a software tool called keyphrase identification program (KIP) for identifying topical concepts from medical documents.  ...  For the noun phrases they disagreed with, a third expert was asked to decide if they were simple noun phrases.  ... 
doi:10.1016/j.jbi.2006.02.001 pmid:16545986 fatcat:mysdq64h55hylduaihta3lgcni

End-to-end Argument Generation System in Debating

Misa Sato, Kohsuke Yanai, Toshinori Miyoshi, Toshihiko Yanase, Makoto Iwayama, Qinghua Sun, Yoshiki Niwa
2015 Proceedings of ACL-IJCNLP 2015 System Demonstrations  
Users can specify a motion such as This house should ban gambling, and a stance on whether the system agrees or disagrees with the motion.  ...  The "value" indicates a topic that is considered as a positive or negative for people or communities, such as health and education.  ...  When users give a "motion" like This house should ban gambling and a "stance" on whether the system should agree or disagree with the motion, the system generates argument scripts in the first constructive  ... 
doi:10.3115/v1/p15-4019 dblp:conf/acl/SatoYMYISN15 fatcat:q2pgrtw375egvhb3i3fihnluxq

Extraction of Keyphrases from Text: Evaluation of Four Algorithms [article]

Peter D. Turney
2002 arXiv   pre-print
For all five document collections, NRC's Extractor yields the best match with the manually generated keyphrases.  ...  The target keyphrases were generated for human readers; they were not tailored for any of the four keyphrase extraction algorithms.  ...  When testing with the three web page corpora, we alternated between the Journal Article model and the Email model, based on the length of the given web page.  ... 
arXiv:cs/0212014v1 fatcat:zvwb4m3vr5d7bkmwkjqpjspzhm

Developing practical automatic metadata assignment and evaluation tools for internet resources

Gordon W. Paynter
2005 Proceedings of the 5th ACM/IEEE-CS joint conference on Digital libraries - JCDL '05  
This paper describes the development of practical automatic metadata assignment tools to support automatic record creation for virtual libraries, metadata repositories and digital libraries, with particular  ...  and labels, and to store this knowledge in a model.  ...  methods, and the use of extracted keyphrases as the basis of scoring has been described elsewhere [15] .  ... 
doi:10.1145/1065385.1065454 dblp:conf/jcdl/Paynter05 fatcat:rctmi2twfnfe3b2bqkisc7rlje

Distributional Contrastive Embedding for Clarification-based Conversational Critiquing

Tianshu Shen, Zheda Mai, Ga Wu, Scott Sanner
2022 Proceedings of the ACM Web Conference 2022  
., suggesting "chicken fingers" when a restaurant with a "kids menu" was the intended preference.  ...  Specifically, we incorporate Distributional Contrastive Embeddings of critiqueable keyphrases with user preference embeddings in a Variational Autoencoder recommendation framework that we term DCE-VAE.  ...  The corresponding Probabilistic Graphical Model is shown in Figure 2(a) . 1 The keyphrases could be extracted from informal descriptions of all items the user interacted with.  ... 
doi:10.1145/3485447.3512114 fatcat:honbjovihrfttfvaoyviox6tyi

Multimedia Summary Generation from Online Conversations: Current Approaches and Future Directions

Enamul Hoque, Giuseppe Carenini
2017 Proceedings of the Workshop on New Frontiers in Summarization  
With the proliferation of Web-based social media, asynchronous conversations have become very common for supporting online communication and collaboration.  ...  We consider combining textual summary with visual representation of conversational data as a promising way of supporting the user in exploring conversations.  ...  After that, topic labeling generates keyphrases to describe each topic segment in the conversation.  ... 
doi:10.18653/v1/w17-4502 dblp:conf/emnlp/HoqueC17 fatcat:2qv47fwskzflxguj2allw3apoi

Breaking Social Media Bubbles for Information Globalization: A Cross-Cultural and Cross-Language User-Centered Sense-Making Approach

Xiaozhong Liu, Daqing He, Dan Wu
2020 Data and Information Management  
The bubble is defined as a state of intellectual isolation; users become more separated from information that disagrees with their viewpoints when they begin to spend more and more time on social media  ...  The idea of keyphrase generation is specifically aimed at extracting the mentions of concepts and entities from the text.  ... 
doi:10.2478/dim-2020-0020 fatcat:pwjaul6vujce7hd5btuedxnyw4


Enamul Hoque, Giuseppe Carenini
2015 Proceedings of the 20th International Conference on Intelligent User Interfaces - IUI '15  
We then integrate this system with interactive visualization techniques to support the user in exploring long conversations, as well as revising the topic model when the current results are not adequate  ...  does not support human-in-the-loop topic model.  ...  revisions Although the initial topic model is more accurate than models generated by traditional methods for non-conversational text [18] , still the extracted topics may not always match the user's  ... 
doi:10.1145/2678025.2701370 dblp:conf/iui/HoqueC15 fatcat:izzlhd37unh5fk5z6c2wspg7aa

A survey on different dimensions for graphical keyword extraction techniques: Issues and Challenges

Muskan Garg
2021 Artificial Intelligence Review  
Different observations over the need to integrate multiple dimensions has open new research directions in the inter-disciplinary field of network science and NLP, applicable to handle streaming data and  ...  It is observed that the Graphical Keyword Extraction Techniques (GKET) use Graph of Words (GoW) in literature for analysis in different dimensions.  ...  A graph-based ranking model is applied to assign a significance score to each topic. Keyphrases are then generated by selecting a candidate from each of the top-ranked topics.  ... 
doi:10.1007/s10462-021-10010-6 pmid:33907346 pmcid:PMC8062621 fatcat:a4boojb4vrebtkoqwn3l66zo7e

The CALO Meeting Assistant System

Gokhan Tur, Andreas Stolcke, Lynn Voss, Stanley Peters, Dilek Hakkani-Tur, John Dowding, Benoit Favre, Raquel Fernandez, Matthew Frampton, Mike Frandsen, Clint Frederickson, Martin Graciarena (+10 others)
2010 IEEE Transactions on Audio, Speech, and Language Processing  
identification and segmentation, question-answer pair identification, action item recognition, decision extraction, and summarization.  ...  This paper presents the CALO-MA architecture and its speech recognition and understanding components, which include real-time and offline speech transcription, dialog act segmentation and tagging, topic  ...  Topic modeling can be seen as two subtasks: • segmentation, dividing the speech data into topically coherent units (the "when" question); • identification, extracting some representation of the topics  ... 
doi:10.1109/tasl.2009.2038810 fatcat:uy43ygwqxvbjzcxuq5rl3mdbua
« Previous Showing results 1 — 15 out of 125 results