Filters








35,485 Hits in 4.8 sec

Incremental TextRank - Automatic Keyword Extraction for Text Streams

Rui Portocarrero Sarmento, Mário Cordeiro, Pavel Brazdil, João Gama
2018 Proceedings of the 20th International Conference on Enterprise Information Systems  
Major improvements are lowest computation times for the processing of the same text data, in a streaming environment, both in sliding window and incremental setups.  ...  In this paper, we present an update to TextRank, a well-known implementation used to do automatic keyword extraction from text, adapted to deal with streams of text.  ...  As future work, we would like to compare this incremental TextRank system with other systems prepared for text streams.  ... 
doi:10.5220/0006639703630370 dblp:conf/iceis/SarmentoCBG18 fatcat:ji3zx4ov7vaw5ekd6il53oirom

Incremental Sparse TFIDF & Incremental Similarity with Bipartite Graphs [article]

Rui Portocarrero Sarmento, Pavel Brazdil
2018 arXiv   pre-print
In this report, we experimented with several concepts regarding text streams analysis.  ...  We tested an implementation of Incremental Sparse TF-IDF (IS-TFIDF) and Incremental Cosine Similarity (ICS) with the use of bipartite graphs.  ...  stream, to check which pairs of documents' changes similarity.  ... 
arXiv:1811.11746v1 fatcat:l4poiarqvvfqrbaol2efi7xmd4

Contextualization for the Organization of Text Documents Streams [article]

Rui Portocarrero Sarmento, Douglas O. Cardoso, João Gama, Pavel Brazdil
2022 arXiv   pre-print
This document shows a case study with developed architectures of a Text Document Stream Organization, using incremental algorithms like Incremental TextRank, and IS-TFIDF.  ...  In this report paper, we present several experiments with some stream analysis methods to explore streams of text documents.  ...  Figure 3 depicts our concept for an improved incremental TextRank. To deal with changing text streams, we modified the original algorithm's procedure.  ... 
arXiv:2206.02632v1 fatcat:7hzwk2fvojb5llkib5nao7ophy

An Improved System for Sentence-level Novelty Detection in Textual Streams [article]

Xinyu Fu, Eugene Ch'ng, Uwe Aickelin, Lanyun Zhang
2016 arXiv   pre-print
We present a novel event detection system based on the Incremental Term Frequency-Inverse Document Frequency (TF-IDF) weighting incorporated with Locality Sensitive Hashing (LSH).  ...  Our system could efficiently and effectively adapt to the changes within the data streams of any new terms with continual updates to the vector space model.  ...  In other words, the incremental TF-IDF weighting approach yields a more accurate identification on the text streams in a novelty detection system when compared to the baseline.  ... 
arXiv:1605.00122v1 fatcat:fhbwtc2pjregvjrgsxeqlcdfl4

An Improved System for Sentence-level Novelty Detection in Textual Streams

Xinyu Fu, Eugene Ch'ng, Uwe Aickelin, Lanyun Zhang
2015 Social Science Research Network  
In other words, the incremental TF-IDF weighting approach yields a more accurate identification on the text streams in a novelty detection system when compared to the baseline.  ...  It is evident from the following tables that the system applied with the incremental TF-IDF weighting scheme better classifies testing streams Although the execution time of the incremental TF-IDF based  ... 
doi:10.2139/ssrn.2828008 fatcat:u2s4xv6wejchzelxe3hwudowbu

Clustering Microtext Streams for Event Identification

Jie Yin
2013 International Joint Conference on Natural Language Processing  
In the online phase, an incremental process is applied to discover base clusters and maintain detailed summary statistics.  ...  The popularity of microblogging systems has resulted in a new form of Web datamicrotext -which is very different from conventional well-written text.  ...  Becker et al. [2011] proposed an incremental clustering approach to group Twitter messages into clusters, which was similar to the method developed for detecting events in streams of text documents [  ... 
dblp:conf/ijcnlp/Yin13 fatcat:5tpgp324wfamnhcs5ogktagqjy

Topic Models over Text Streams: A Study of Batch and Online Unsupervised Learning [chapter]

Arindam Banerjee, Sugato Basu
2007 Proceedings of the 2007 SIAM International Conference on Data Mining  
Finally, we propose a practical heuristic for hybrid topic modeling, which learns online topic models on streaming text and intermittently runs batch topic models on aggregated documents offline.  ...  Topic modeling techniques have widespread use in text data mining applications. Some applications use batch models, which perform clustering on the document collection in aggregate.  ...  Acknowledgments We would like to thank Jiye Yu for helping with data collection, Misha Bilenko for valuable feedback, Charles Elkan and Tom Griffiths for providing code for the batch EDCM and LDA models  ... 
doi:10.1137/1.9781611972771.40 dblp:conf/sdm/BanerjeeB07 fatcat:67c7j37ckbbhvfs2yutspgz6ii

Collecting Valuable Information from Fast Text Streams [chapter]

Baoyuan Qi, Gang Ma, Zhongzhi Shi, Wei Wang
2014 IFIP Advances in Information and Communication Technology  
It has become a challenging work to collect valuable information from fast text streams. In this work, we propose a method which gains useful information effectively and efficiently.  ...  The experimental results show that it has the strong adaption ability, low latency and high quality support for the complex query combination compared with the conventional methods.  ...  Introduction Text streams is a kind of data stream, which is composed of texts, such as news feed, blog, weibo, etc..  ... 
doi:10.1007/978-3-662-44980-6_11 fatcat:obxmgq5hu5fk5fvnue4y7qhxtm

Incremental clustering for profile maintenance in information gathering web agents

Gabriel L. Somlo, Adele E. Howe
2001 Proceedings of the fifth international conference on Autonomous agents - AGENTS '01  
We compare a manual profile maintenance technique in which the user supplies the document topic, and two incremental clustering methods (greedy and the doubling algorithm) for automated maintenance of  ...  user behavior and thus speed up data collection, exert additional experimental control and improve the objectivity of our results.  ...  In contrast, text filtering makes binary decisions of whether or not to disseminate items from a continuously incoming stream of documents [5] ; the decision is made based on whether a document adequately  ... 
doi:10.1145/375735.376306 dblp:conf/agents/SomloH01 fatcat:ynuqzpp62jhejkryaqzl7xyfza

Improved Document Clustering Technique Using K-Mean Method

Purvi Khare
2016 International Journal of Scientific Research and Management  
Clustering of document is important for the purpose of document organization, summarization, topic extraction and information retrieval in an efficient way.  ...  In this paper, we are providing a methodology for more accurate document clustering.  ...  of text documents available through World Wide Web and corporate document management systems.  ... 
doi:10.18535/ijsrm/v4i1.02 fatcat:ozcved2cb5awbd3uvjy3zoyucy

Estimating timestamp from incomplete news corpus

Takao Miura, Isamu Shioya, Hiroshi Uejima
2004 Communications in Information and Systems  
Recently there have been a lot of researches for summarizing news stream and for detecting edges of new events in the news stream.  ...  Here we learn temporal information and topic information by means of both EM algorithm and incremental clustering, then we estimate timestamp of news article based on events that are discussed in news  ...  According to this property, we will estimate timestamp of documents accurately based on events in documents. That's why we discuss incremental clustering techniques for document stream.  ... 
doi:10.4310/cis.2004.v4.n4.a1 fatcat:ygnygx27irhtjbff66fo2jnrlm

Incremental visual text analytics of news story development

Milos Krstajic, Mohammad Najm-Araghi, Florian Mansmann, Daniel A. Keim, Pak Chung Wong, David L. Kao, Ming C. Hao, Chaomei Chen, Robert Kosara, Mark A. Livingston, Jinah Park, Ian Roberts
2012 Visualization and Data Analysis 2012  
To demonstrate the usefulness of our system, case studies with real news data are presented and show the capabilities for detailed dynamic text stream exploration.  ...  We employ text clustering techniques to automatically extract stories from online news streams and present a visualization that: 1) shows temporal characteristics of stories in different time frames with  ...  Our research efforts will continue in the direction of integrating incremental text analysis with novel visualization methods that will enable information analysts to analyze and understand growing document  ... 
doi:10.1117/12.912456 dblp:conf/vda/KrstajicNMK12 fatcat:36iw4sh5r5dvxbqpwr6knjh7hy

Mining Text Streams [chapter]

Charu C. Aggarwal
2012 Mining Text Data  
An example in the latter category would be the massive text streams created by news-wire services.  ...  Such text streams provide unprecedented challenges to data mining algorithms from an efficiency perspective.  ...  A method proposed in [11] is quite similar to that proposed in [47] , except that it proposes a number of improvements in how the tf-idf model is incrementally maintained for computation of similarity  ... 
doi:10.1007/978-1-4614-3223-4_9 fatcat:x3q4nx36zzea7pn2y5w7xmli34

A Framework for Fast Polarity Labelling of Massive Data Streams [article]

Huilin Wu and Mian Lu and Zhao Zheng and Shuhao Zhang
2022 arXiv   pre-print
We address the associated implementation challenges and propose a list of techniques including both algorithmic improvements and system optimizations.  ...  data streams (almost 16,000 tuples/sec) without any manual efforts.  ...  We have presented PLStream, a new framework designed for annotating fast ongoing unlabelled data streams at scale on modern parallel machines.  ... 
arXiv:2203.12368v1 fatcat:6v7hipqz7ngh3b5b2fta7aeq2q

Evaluation of Neural Network Classification Systems on Document Stream [article]

Joris Voerman, Aurelie Joseph, Mickael Coustaty, Vincent Poulain d Andecy, Jean-Marc Ogier
2020 arXiv   pre-print
NN-Based document classification systems need to be adapted to resolve these two problems before they can be considered for use in a company document stream.  ...  In this paper, we analyse the efficiency of NN-based document classification systems in a sub-optimal training case, based on the situation of a company document stream.  ...  With this distribution we can finally simulate results of neural network methods on a document stream with an ideal test set and evaluate the global impact of document streams on neural network method  ... 
arXiv:2007.07547v1 fatcat:cwpu2fnnkjh67ilsksgjf7kj4e
« Previous Showing results 1 — 15 out of 35,485 results