78,295 Hits in 5.0 sec

The maximum-margin approach to learning text classifiers: methods theory, and algorithms

Thorsten Joachims
2001 Jahrestagung der Gesellschaft für Informatik  
Die Dissertation zeigt den Zusammenhang dieser Maße mit den statistischen Eigenschaften von Text, deren Umsetzung in effektiven und praktikablen Methoden zur Textklassifikation und ihre Implementierung  ...  . £ Erschienen als: Learning to Classify Text Using Support Vector Machines, Kluwer, 2002.  ...  I can't I need to get the specs, or at least a for QuickTime.  ... 
dblp:conf/gi/Joachims01 fatcat:pkvinnvyrfbybehz472mdimjka

Representative Sampling for Text Classification Using Support Vector Machines [chapter]

Zhao Xu, Kai Yu, Volker Tresp, Xiaowei Xu, Jizhi Wang
2003 Lecture Notes in Computer Science  
In order to reduce human efforts, there has been increasing interest in applying active learning for training text classifiers.  ...  This paper describes a straightforward active learning heuristic, representative sampling, which explores the clustering structure of 'uncertain' documents and identifies the representative samples to  ...  Then the maximum margin is interpreted as the maximum distance from the point w to restricting boundaries in W, which correspond support vectors in X.  ... 
doi:10.1007/3-540-36618-0_28 fatcat:qrlfooa2nfefnfenwemffijww4

Machine Learning Approach for Text Summarization

Amita Arora, Akanksha Diwedy, Manjeet Singh, Naresh Chauhan
2017 International Journal of Database Theory and Application  
With the abundance of interminable text documents, providing summaries can help in retrieval of relevant information very quickly.  ...  The technique is to extract those sentences from the document that contain important information.  ...  Only positively classified sentences are taken, and they are ranked according to the their distance from the maximum margin hyperplane.  ... 
doi:10.14257/ijdta.2017.10.8.08 fatcat:ruv62vfbtjfo7lybbhx3hykmam

Mining the Web for Relations between Digital Devices using a Probabilistic Maximum Margin Model

Oksana Yakhnenko, Barbara Rosario
2008 International Joint Conference on Natural Language Processing  
Our results show that the maximum margin model consistently outperforms the other methods.  ...  We use a Naïve Bayes model trained to maximize the margin and compare its performance with several other comparable methods.  ...  Acknowledgment The authors would like to thank the reviewers for their feedback and comments; William Schilit for invaluable insight and help and for first suggesting using the MTurk to gather labeled  ... 
dblp:conf/ijcnlp/YakhnenkoR08 fatcat:mgrojkvse5calnzxa3p7sucahy

Classification of Users' Opinions and Posts on Facebook Using Machine Learning Approaches

Ibrahim sayed, Mohamed Nour, Mohammed Badawy, Ehsan Abed
2022 Menoufia Journal of Electronic Engineering Research  
Some criteria are considered to evaluate the performance of the classification process mainly: precision, recall, F-measure, and learning time.  ...  The reviews are collected from the Arabic Facebook posts. Several experiments are done to evaluate the performance of the adopted classifiers.  ...  ANALYSIS OF SOME APPROACHES FOR CLASSIFYING ARABIC TEXT There are several types of classifiers to classify and/or categorize the different documents.  ... 
doi:10.21608/mjeer.2022.79630.1037 fatcat:hvqxjyjc6vdzdealrasovdaeny

Sentiment Classification using Machine Learning Techniques

2016 International Journal of Science and Research (IJSR)  
Specifically, we compared two supervised machine learning approaches SVM, Navie Bayes for Sentiment Classification of Reviews. Results states that Naïve Bayes approach outperformed the svm.  ...  The opinions obtained from those can be classified in to positive or negative which can be used by customer to make product choice and by businessmen for finding customer satisfaction .This paper studies  ...  The basic idea behind SVM classification is to find hyper-plane with maximum margin that separates the document vector in one class from the other with maximum margin.  ... 
doi:10.21275/v5i4.nov162724 fatcat:gprisgvgu5bzjpdyqillll6doy

UB at TREC 11: Batch and Adaptive Filtering

Munirathnam Srikanth, Xiaoyun Wu, Rohini K. Srihari
2002 Text Retrieval Conference  
Compare to ERM, SRM is more suited when the training data set is limited. SVM, the simplest linear form of SRM, is nothing but a maximum margin linear classifier.  ...  The maximum margin hyperplane given training set S is thus defined as the hyperplane with respect to which the training set has maximum geometric margin. There are two major advantages of SVM.  ... 
dblp:conf/trec/SrikanthWS02 fatcat:kbgjm6ptvjbrrjebndnsf7dohy

Active Learning Strategies for Technology Assisted Sensitivity Review [chapter]

Graham McDonald, Craig Macdonald, Iadh Ounis
2018 Lecture Notes in Computer Science  
Moreover, this approach results in a 51% reduction in the number of documents required to be reviewed to achieve the same level of classification accuracy, compared to when the approach is deployed without  ...  Moreover, to effectively assist sensitivity review, such assistive technologies must incorporate reviewer feedback to enable sensitivity classifiers to quickly learn and adapt to the sensitivities within  ...  s approach was designed to rank documents in an order that would achieve the maximum increase in overall classification if a reviewer was to start from the top of the ranking and proceed down the list  ... 
doi:10.1007/978-3-319-76941-7_33 fatcat:kroy3sx4czef5g5fuo7k45q6vq

On Text-based Mining with Active Learning and Background Knowledge Using SVM

Catarina Silva, Bernardete Ribeiro
2006 Soft Computing - A Fusion of Foundations, Methodologies and Applications  
Text mining, intelligent text analysis, text data mining and knowledge-discovery in text are generally used aliases to the process of extracting relevant and non-trivial information from text.  ...  Besides studying the influence of several pre-processing methods and concluding on their relative significance, we also evaluate the benefits of introducing background knowledge in a SVM text classifier  ...  The optimal separating hyperplane is defined as the one giving the maximum margin between the training examples that are closest to the hyperplane.  ... 
doi:10.1007/s00500-006-0080-8 fatcat:4o23dx2xzvefjjlh7okbkfdpme

Analysis of Various Sentiment Classification Techniques

Vimalkumar B., Bhumika M.
2016 International Journal of Computer Applications  
The main aim of this paper is to find out approaches that generate output with good accuracy.  ...  Sentiment analysis is an ongoing research area in the field of text mining.  ...  Most widely approaches have been discussed below. SVM Support vector machine examines the data, identify hyper plane that classify data in to two classes with maximum margin.  ... 
doi:10.5120/ijca2016909259 fatcat:r32jcbc26zg5lc7qj6mzenmk4e

Question Classification using Machine Learning Approaches

Arun DPanicker, Athira U, Sreesha Venkitakrishnan
2012 International Journal of Computer Applications  
Here we are extending the previous methods for text categorization to question categorization and making a comparative study of the performance of two approaches, Naïve Bayes and Support Vector Machine  ...  Many approaches to question classification have been proposed and have achieved reasonable results. The dominant approaches are machine learning and context based classification.  ...  Here we are proposing two approaches, SVM and Naïve Bayes, which have been previously used for text classification, for network-based learning wherein the questions posted by students on-line will be classified  ... 
doi:10.5120/7405-0101 fatcat:i75mbt63nrdpvduvbmlr4k5654

Sentiment Analysis by using deep learning and Machine learning Techniques: A Review

2021 International Journal of Advanced Trends in Computer Science and Engineering  
The gap in the precision of these two approaches, however, is not as important enough in certain situations, and so it is best to apply and use the machine learning approaches and methods because these  ...  Deep learning is the part of machine learning and deals with the algorithm, which is most widely used as Neural network, neural belief, etc., in which neuronal implementations are considered.  ...  82.5 to 85% Decision trees classifiers [15] 85 to 90% Maximum Entropy classifier (ME) [14] 79 to 83.25% Support vector machine (Chinese texts) [9] 76.47% Naive Bayes (Chinese texts) [6] 74.13%  ... 
doi:10.30534/ijatcse/2021/421022021 fatcat:sce5t4wdkrclfjmtz4vmlrdue4

Movie Review analysis using Rule-Based &Support Vector Machines methods

Swati A. Kawathekar
2012 IOSR Journal of Engineering  
Sentiment analysis (SA) is broad forte of Natural language processing which deals with the computational treatment of opinion, sentiment and subjectivity in text.  ...  This paper combines rule-based classification, supervised learning and machine learning into a new combined method. This method is tested on movie review.  ...  margin the lower the generalization error of the classifier.  ... 
doi:10.9790/3021-0203389391 fatcat:zinatr6db5cx7bgw3mpki2llcu

A stopping criterion for active learning

Andreas Vlachos
2008 Computer Speech and Language  
The statistical learning models used in our study are support vector machines (SVMs), maximum entropy models and Bayesian logistic regression and the tasks performed are text classification, named entity  ...  Active learning (AL) is a framework that attempts to reduce the cost of annotating training material for statistical learning methods.  ...  The author would like to thank Mark Craven, Zoubin Ghahramani, Bobby Grammacy and the anonymous reviewers for helpful discussion and comments on this work.  ... 
doi:10.1016/j.csl.2007.12.001 fatcat:snkui4pcp5dbxdxqndmpzl7zvm

A Comparison of Text-Categorization Methods Applied to N-Gram Frequency Statistics [chapter]

Helmut Berger, Dieter Merkl
2004 Lecture Notes in Computer Science  
Furthermore the impact of using available e-mail specific meta-information on classification performance is explored and the findings are presented.  ...  In particular, a Naïve Bayes classifier [7] , partial decision trees (PART) as a rule learning approach [8] and support vector machines trained with the sequential minimal optimization algorithm [9  ...  One objective of this study is to determine the influence of different document representations on the performance of different text-classification approaches.  ... 
doi:10.1007/978-3-540-30549-1_92 fatcat:baobvepejzb77p3i5zcl65rk6m
« Previous Showing results 1 — 15 out of 78,295 results