OntoSeg: A Novel Approach to Text Segmentation Using Ontological Similarity

Bayomi, Mostafa; Levacher, Killian; Ghorab, M. Rami; Lawless, Séamus

doi:10.1109/icdmw.2015.6

Cited by 13 publications

(10 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Later, supervised methods included topic models (Riedl and Biemann, 2012) by calculating a coherence score using dense topic vectors obtained by LDA. Bayomi et al (2015) exploited ontologies to measure semantic similarity between text blocks. Alemi and Ginsparg (2015) and Naili et al (2017) studied how word embeddings can improve classical segmentation approaches.…”

Section: Related Workmentioning

confidence: 99%

SECTOR: A Neural Model for Coherent Topic Segmentation and Classification

Arnold

Schneider

Cudré-Mauroux

et al. 2019

Transactions of the Association for Computational Linguistics

View full text Add to dashboard Cite

When searching for information, a human reader first glances over a document, spots relevant sections and then focuses on a few sentences for resolving her intention. However, the high variance of document structure complicates to identify the salient topic of a given section at a glance. To tackle this challenge, we present SECTOR, a model to support machine reading systems by segmenting documents into coherent sections and assigning topic labels to each section. Our deep neural network architecture learns a latent topic embedding over the course of a document. This can be leveraged to classify local topics from plain text and segment a document at topic shifts. In addition, we contribute WikiSection, a publicly available dataset with 242k labeled sections in English and German from two distinct domains: diseases and cities. From our extensive evaluation of 20 architectures, we report a highest score of 71.6% F1 for the segmentation and classification of 30 topics from the English city domain, scored by our SECTOR LSTM model with bloom filter embeddings and bidirectional segmentation. This is a significant improvement of 29.5 points F1 compared to state-of-the-art CNN classifiers with baseline segmentation. 1 Our source code is available under the Apache License 2.0 at https

show abstract

Section: Related Workmentioning

confidence: 99%

SECTOR: A Neural Model for Coherent Topic Segmentation and Classification

Arnold

Schneider

Cudré-Mauroux

et al. 2019

Transactions of the Association for Computational Linguistics

View full text Add to dashboard Cite

show abstract

“…This causes generating the segment boundaries with the disadvantage of words being repeated throughout the process of segmentation as reported in the existing works [35]. To overcome this issue, different approaches were also proposed [36, 37], but they do not rely on a training phase or directly applied to text data. However, existing works found comparative results using ontological similarity in conjunction with labelled data, but their document representation assume each text unit as single piece of information as well as their thematic information is being lost, and thus, segments might not be related or labelled with any topical information [38, 39].…”

Section: Related Workmentioning

confidence: 99%

“…segments), which is proclaimed as ‘speaking about’ for a specific topic. However, traditional approaches [37–41] find topics underlie on boundaries by default: either they consider topical borderline is to be sketched in the perspective of no man’s land between two distinguished topical areas, or a large variance in the vocabulary could occur. Furthermore, these approaches limit the data to words rather using topics and they are based on similarity or density measure, thus, letting go to discourse and semantic information contained in the text units.…”

Section: Document Representation and Scoringmentioning

confidence: 99%

An ensemble clustering approach for topic discovery using implicit text segmentation

Memon

Chen

et al. 2020

Journal of Information Science

View full text Add to dashboard Cite

Text segmentation (TS) is the process of dividing multi-topic text collections into cohesive segments using topic boundaries. Similarly, text clustering has been renowned as a major concern when it comes to multi-topic text collections, as they are distinguished by sub-topic structure and their contents are not associated with each other. Existing clustering approaches follow the TS method which relies on word frequencies and may not be suitable to cluster multi-topic text collections. In this work, we propose a new ensemble clustering approach (ECA) is a novel topic-modelling-based clustering approach, which induces the combination of TS and text clustering. We improvised a LDA-onto (LDA-ontology) is a TS-based model, which presents a deterioration of a document into segments (i.e. sub-documents), wherein each sub-document is associated with exactly one sub-topic. We deal with the problem of clustering when it comes to a document that is intrinsically related to various topics and its topical structure is missing. ECA is tested through well-known datasets in order to provide a comprehensive presentation and validation of clustering algorithms using LDA-onto. ECA exhibits the semantic relations of keywords in sub-documents and resultant clusters belong to original documents that they contain. Moreover, present research sheds the light on clustering performances and it indicates that there is no difference over performances (in terms of F-measure) when the number of topics changes. Our findings give above par results in order to analyse the problem of text clustering in a broader spectrum without applying dimension reduction techniques over high sparse data. Specifically, ECA provides an efficient and significant framework than the traditional and segment-based approach, such that achieved results are statistically significant with an average improvement of over 10.2%. For the most part, proposed framework can be evaluated in applications where meaningful data retrieval is useful, such as document summarization, text retrieval, novelty and topic detection.

show abstract

“…The domains were selected at random and the abstracts were randomly chosen from within the following domains: Accidents, Natural Disasters, Politics, Famous People, Sports, and Animals. Two summaries were generated for each abstract, one summary was generated without applying Anaphora Resolution and the second summary was generated by applying Anaphora Resolution before summarisation 1 . When we tried to generate the two summaries for each article, we noticed in 45 articles that the two summaries were identical, which reflected the possibility that AR may have had no impact on the produced summaries.…”

Section: Datasetmentioning

confidence: 99%

“…Natural Language Processing (NLP) has different tasks [1,2]. One of these tasks is Automatic Text Summarisation (TS) that has been the subject of a lot of interest in the NLP community in recent years [2].…”

Section: Introductionmentioning

confidence: 99%

Towards Evaluating the Impact of Anaphora Resolution on Text Summarisation from a Human Perspective

Bayomi

Levacher

Ghorab

et al. 2016

Natural Language Processing and Information Systems

Self Cite

View full text Add to dashboard Cite

Abstract. Automatic Text Summarisation (TS) is the process of abstracting key content from information sources. Previous research attempted to combine diverse NLP techniques to improve the quality of the produced summaries. The study reported in this paper seeks to establish whether Anaphora Resolution (AR) can improve the quality of generated summaries, and to assess whether AR has the same impact on text from different subject domains. Summarisation evaluation is critical to the development of automatic summarisation systems. Previous studies have evaluated their summaries using automatic techniques. However, automatic techniques lack the ability to evaluate certain factors which are better quantified by human beings. In this paper the summaries are evaluated via human judgment, where the following factors are taken into consideration: informativeness, readability and understandability, conciseness, and the overall quality of the summary. Overall, the results of this study depict a pattern of slight but not significant increases in the quality of summaries produced using AR. At a subject domain level, however, the results demonstrate that the contribution of AR towards TS is domain dependent and for some domains it has a statistically significant impact on TS.

show abstract

OntoSeg: A Novel Approach to Text Segmentation Using Ontological Similarity

Cited by 13 publications

References 29 publications

SECTOR: A Neural Model for Coherent Topic Segmentation and Classification

SECTOR: A Neural Model for Coherent Topic Segmentation and Classification

An ensemble clustering approach for topic discovery using implicit text segmentation

Towards Evaluating the Impact of Anaphora Resolution on Text Summarisation from a Human Perspective

Contact Info

Product

Resources

About