Similarity-Based Query-Focused Multi-document Summarization Using Crowdsourced and Manually-built Lexical-Semantic Resources

2016 IEEE Second International Conference on Big Data Computing Service and Applications (BigDataService)

Oussalah

2016

Self Cite

This paper proposes an innovative graph-based text summarization model for generic single and multi-document summarization. The approach involves four unique processing stages: parsing sentences semantically using Semantic Role Labeling (SRL), grouping semantic arguments while matching semantic roles to Wikipedia concepts, constructing a weighted semantic graph for each document and linking its sentences (nodes) through the semantic relatedness of the Wikipedia concepts. An iterative ranking algorithm is then applied to the document graphs to extract the most important sentences deemed as the summary. The empirical evaluation of the proposed summarization model on a standard dataset from the Document Understanding Conference (DUC) showed the effectiveness of the approach which outperformed the baseline comparators in terms of ROUGE scores.

Section: Iterative Sentence Ranking For Summary Extractionmentioning

confidence: 99%

Section: Iterative Sentence Ranking For Summary Extractionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Iterative Graph-Based Generic Single and Multi Document Summarization Approach Using Semantic Role Labeling and Wikipedia Concepts

2016 IEEE Second International Conference on Big Data Computing Service and Applications (BigDataService)

Oussalah

2016

Self Cite

“…On the other hand, data analytics is the technical process of transforming raw data into meaningful insights for better decision making. This includes text analytics, e.g., summarization, since unstructured text forms the highest percentage of current big data [11][12][13][14].…”

Section: Introductionmentioning

confidence: 99%

Trends of digitalization and adoption of big data & analytics among UK SMEs: Analysis and lessons drawn from a case study of 53 SMEs

2020 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC)

Weber

2020

Small and Medium Enterprises (SMEs) now generate digital data at an unprecedented rate from online transactions, social media marketing and associated customer interactions, online product/service reviews and feedback, clinical diagnosis, Internet of Things (IoT) sensors, and production processes. All these forms of data can be transformed into monetary value if put into a proper data value chain. This requires both skills and IT investments for the longterm benefit of businesses. However, such spending is beyond the capacity of most SMEs due to their limited resources and restricted access to finance. This paper presents lessons learned from a case study of 53 UK SMEs, mostly from the West Midlands region of England, supported as part of a 3-year ERDF 1 project -Big Data Corridor 2 -in the areas of big data management, analytics and related IT issues. Based on our study's sample companies, several perspectives including digital technology trends, challenges facing the UK SMEs, and the state of their adoption in data analytics and big data, are presented.

“…It classifies the semantic roles of syntactic arguments within a given frame of the sentence and with respect to the predicate. The use of the proposed SRL-ESA Wikipedia graph-based generic summarization model is motivated by the following: the successful application of Wikipedia-based metric to the tasks of named entity semantic relatedness and query-focused summarization in our previous work [3,4]; the high lexical coverage of Wikipedia which led to its popularity as a reliable lexical resource for different NLP tasks such as semantic representation [6], word semantic similarity [5] and text classification [7]; the status quo in which current knowledge-based summarization methods underestimate the importance of sentence syntactic order and semantic roles, consequently undermining an accurate similarity computation and hence leading to poor scoring functions for summary extraction; the computation of word similarities in isolation from the surrounding context, thus ignoring significant semantic information conveyed by these words if associated with their roles when analysing them semantically. The chief contributions of this paper are:…”

Section: Introductionmentioning

confidence: 99%

SRL-ESA-TextSum: A text summarization approach based on semantic role labeling and explicit semantic analysis

Information Processing & Management

Oussalah

2019

Self Cite

Automatic text summarization attempts to provide an effective solution to today's unprecedented growth of textual data. This paper proposes an innovative graph-based text summarization framework for generic single and multi document summarization. The summarizer benefits from two well-established text semantic representation techniques; Semantic Role Labelling (SRL) and Explicit Semantic Analysis (ESA) as well as the constantly evolving collective human knowledge in Wikipedia. The SRL and ESA methods are used to achieve sentence semantic parsing and to represent its words as a vector of weighted Wikipedia concepts, respectively. The essence of the developed framework is to construct a unique concept graph representation underpinned by semantic role-based multi-node (under sentence level) vertices for summarization. We have empirically evaluated the summarization system using the standard publicly available dataset from Document Understanding Conference 2002 (DUC2002). Experimental results indicate that the proposed summarizer outperforms all state-of-the-art related comparators in the single document summarization based on the ROUGE-1 and ROUGE-2 measures, while also ranking second in the ROUGE-1 and ROUGE-SU4 scores for the multi-document summarization. Results also tell that the system is scalable, i.e., varying the evaluation data size is shown to have little impact on the summarizer performance, particularly for the single document summarization task. In a nutshell, the findings demonstrate the power of the role-based and vectoral semantic representation when combined with the crowd-sourced knowledge base in Wikipedia.