Text Summarization and Singular Value Decomposition

Tailored trusted spaces (TTS) need to identify data sensitive to user context and ensure sensitive data is not leaked. This task is similar to Data Leak Prevention (DLP) solutions, which monitor and control corporate data flow. Current DLP techniques find data that matches user defined syntactic patterns [1,2]. However, user context is usually described by information semantics, rather than data syntax. In this paper, we propose a method which extracts semantic features from a small number of training documents describing user context, and use the presence of these semantic features to decide whether an input document contains sensitive information or not. Test results from three document sets show the proposed method giving 96%, 90%, 75% true positive and 3%, 14%, 26% false positive rates.

show abstract

“…The magnitude of singular values in S k reflects the influence of concepts [10]. Figure 3 plots the magnitude of singular values in S k .…”

Section: Influence Of Conceptsmentioning

confidence: 98%

Tailoring trusted semantic information

Zhao

Brooks

2013

Proceedings of the Eighth Annual Cyber Security and Information Intelligence Research Workshop

View full text Add to dashboard Cite

show abstract

“…Extraction using dimension reduction is another summarization technique based on SVD (Hirohata et al, 2003;Steinberger and Ježek, 2004;Murray et al, 2005a,b). As shown in Fig.…”

Section: Extraction Using Dimension Reduction Based On Svdmentioning

confidence: 99%

Sentence-extractive automatic speech summarization and evaluation techniques

Hirohata

Shinnaka

Iwano

et al. 2006

Speech Communication

View full text Add to dashboard Cite

“…Each concept (topic) is represented in the summary by a sentence, which captures it the best. However, Steinberger and Ježek [10] showed that this approach fails to include into the summary sentences, which capture many concepts well, but have the highest score for none of them. They proposed a modification of the selection of sentences; sentences are selected based on their overall score computed as a combination of scores for each concept (topic).…”

Section: Related Workmentioning

confidence: 99%

“…As the final step, we select sentences with the highest score computed by a method proposed in [10] (sentences are selected from the original, not the translated document):…”

Section: Sentences Selectionmentioning

confidence: 99%

Personalized Text Summarization Based on Important Terms Identification

Móro

2012

2012 23rd International Workshop on Database and Expert Systems Applications

View full text Add to dashboard Cite

Abstract-Automatic text summarization aims to address the information overload problem by extracting the most important information from a document, which can help a reader to decide whether it is relevant or not. In this paper we propose a method of personalized text summarization which improves the conventional automatic text summarization methods by taking into account the differences in readers' characteristics. We use annotations added by readers as one of the sources of personalization. We have experimentally evaluated the proposed method in the domain of learning, obtaining better summaries capable of extracting important concepts explained in the document when considering the relevant domain terms in the process of summarization.

show abstract

Text Summarization and Singular Value Decomposition

Cited by 53 publications

References 6 publications

Tailoring trusted semantic information

Tailoring trusted semantic information

Sentence-extractive automatic speech summarization and evaluation techniques

Personalized Text Summarization Based on Important Terms Identification

Contact Info

Product

Resources

About