Intelligent Extractive Text Summarization Using Fuzzy Inference Systems

Kiani, Behnam; Akbarzadeh, T.; Moeinzadeh, M-Hossein

doi:10.1109/iceis.2006.1703156

Cited by 3 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In training the logistic regression model with the sentence length and lexical features, we found that lengthy sentences often contained useful information needed to create a good summary. This is consistent with the literature which states that the sentence length is used to eliminate short sentences from the summary that do not contain useful information, such as author names or code extractions [9]- [11]. However, we also found that only using the length of the sentence was not enough to capture all of the useful sentences, as selecting longer sentences resulted in quickly reaching the word percentage threshold.…”

Section: Discussionsupporting

confidence: 90%

Feature Evaluation for Automatic Bug Report Summarization (S)

Galappaththi¹,

Anvik²

2019

International Conferences on Software Engineering and Knowledge Engineering

View full text Add to dashboard Cite

Bug reports can be lengthy due to long descriptions and long conversation threads. Automatic summarization of the text in a bug report can reduce the time spent by software project members on understanding the content of a bug report. Our work further examines Rastkar et al.'s use of a logistic regression model to determine which sentences from the text of a bug report should be extracted for creating a summary. Using their publicly available bug report corpus, which contains manually annotated bug reports, we examined two aspects regarding the features used by the model. First, we examined how much of a reduction occurs in the precision and recall if some of the more complex features are not used. Second, we examined how the use of different feature combinations affects the precision and recall of the models. We found that the absence of some of the complex features resulted in a modest decrease in precision and recall, and confirmed that some features, such as sentence length, were the most significant features for bug report summarization.

show abstract

Section: Discussionsupporting

confidence: 90%

Feature Evaluation for Automatic Bug Report Summarization (S)

Galappaththi¹,

Anvik²

2019

International Conferences on Software Engineering and Knowledge Engineering

View full text Add to dashboard Cite

show abstract

“…This feature can involve several items, such as the position of a sentence in the document as a whole, its the position in a section, in a paragraph, etc., and has presented good results in several research projects (Liu, 2011;Kiani, 2002;Kiyomarsi, 2011;Neto, 2002;Duck, 2006). We use here the percentile of the sentence position in the document, as proposed by Nevill-Manning (Mani, 1999); the final value is normalized to take on values between 0 and 1.…”

Section: Automatic Text Summarization Using a Machine Learning Approachmentioning

confidence: 86%

“…Since the seminal work of Luhn (Kiyomarsi, 2011) text processing tasks frequently use features based on IR measures (liu, 200;Kiani, 2002;Duc, 2006), In the context of IR, some very important measures are term frequency (TF) and term frequency -IDF) (Jones, 1999). In text summarization we can employ the same idea: in this case we have a single document d, and we have to select a set of relevant sentences to be included in the extractive summary out of all sentences in d. Hence, the notion of a collection of documents in IR can be replaced by the notion of a single document in text summarization.…”

Section: Automatic Text Summarization Using a Machine Learning Approachmentioning

confidence: 99%

“…This feature is employed to penalize sentences that are too short, since these sentences are not expected to belong to the summary (Kiani, 2002) We use the normalized length of the sentence, which is the ratio of the number of words occurring in the sentence over the number of words occurring in the longest sentence of the document.…”

Section: Automatic Text Summarization Using a Machine Learning Approachmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of Automatic Text Summarizations based on Human Summaries

Kiyoumarsi

2015

Procedia - Social and Behavioral Sciences

View full text Add to dashboard Cite

The goal of this paper is to compare summaries generated by different automatic text summarization methods and those generated by human beings. To achieve this end, we did two series of experiments: in the first one, we employed automatically produced extractive summaries; in the second one, manually-produced summaries obtained by several English teachers were used. Our automatic summaries were obtained using Fuzzy method and Vector approach. Using Rouge evaluation system, we compared the manually-produced summaries and the automatically-produced ones. Rouge evaluation of generated summaries indicated the superiority of summaries produced by humans over the automatically produced summaries. On the other hand, the comparison between the generated summaries showed that summaries produced by Fuzzy method were much more acceptable and understandable compared to summaries produced by Vector approach. This can provide support for the replacement of manually generated summaries by summaries produced using Fuzzy method in certain cases where real time summaries are needed.

show abstract

Enhanced Feature-Based Automatic Text Summarization SystemUsingSupervised Technique

Ali

Al-Dahoud

Hawashin

2016

IJCT

View full text Add to dashboard Cite

In this work, we propose an efficient text summarization methodby ranking sentences according to their scores that use a combination of existing and improved sentence features.Â Many works in the literature proposed improvements to text summarization but this field still needs more improvement. For this purpose, we propose improvements to Sentence position, Sentence length, and Key wordsentence features. Afterwards, we find the optimal combination between these features and some existing features such as Term frequency, Sentence centrality, Title similarity, and Upper case of word. By usingmachine learning techniques, mainly SVM, Naive Bayes and Decision Tree classifiersour paper evaluates two feature groups: a combination of seven features without any improvements,and the same seven features after making some improvements onSentence position, Sentence length, and Key word sentence features to enhance the performance of text summarization system.Experimental results showed that making enhancements on some features improved the accuracy.

show abstract

Intelligent Extractive Text Summarization Using Fuzzy Inference Systems

Cited by 3 publications

References 9 publications

Feature Evaluation for Automatic Bug Report Summarization (S)

Feature Evaluation for Automatic Bug Report Summarization (S)

Evaluation of Automatic Text Summarizations based on Human Summaries

Enhanced Feature-Based Automatic Text Summarization SystemUsingSupervised Technique

Contact Info

Product

Resources

About