Classification of RSS feed news items using ontology

Agarwal, Shikha; Singhal, Archana; Bedi, Punam

doi:10.1109/isda.2012.6416587

Cited by 10 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If an exact/partial match is found, the terms are replaced by the associated concepts forming a richer semantic representation of the document. In Reference [16], document related concepts are fetched from the domain ontology forming a CF-IDF (Concept Frequency-Inverse Document Frequency) to classify news-feeds. In Reference [17], news documents are clustered by measuring similarity between named entities from a document and related matched information from DBPedia (https://www.dbpedia.org/, accessed on 10 May 2021.)…”

Section: Related Workmentioning

confidence: 99%

Linked Data Triples Enhance Document Relevance Classification

et al. 2021

View full text Add to dashboard Cite

Standardized approaches to relevance classification in information retrieval use generative statistical models to identify the presence or absence of certain topics that might make a document relevant to the searcher. These approaches have been used to better predict relevance on the basis of what the document is “about”, rather than a simple-minded analysis of the bag of words contained within the document. In more recent times, this idea has been extended by using pre-trained deep learning models and text representations, such as GloVe or BERT. These use an external corpus as a knowledge-base that conditions the model to help predict what a document is about. This paper adopts a hybrid approach that leverages the structure of knowledge embedded in a corpus. In particular, the paper reports on experiments where linked data triples (subject-predicate-object), constructed from natural language elements are derived from deep learning. These are evaluated as additional latent semantic features for a relevant document classifier in a customized news-feed website. The research is a synthesis of current thinking in deep learning models in NLP and information retrieval and the predicate structure used in semantic web research. Our experiments indicate that linked data triples increased the F-score of the baseline GloVe representations by 6% and show significant improvement over state-of-the art models, like BERT. The findings are tested and empirically validated on an experimental dataset and on two standardized pre-classified news sources, namely the Reuters and 20 News groups datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Linked Data Triples Enhance Document Relevance Classification

et al. 2021

View full text Add to dashboard Cite

show abstract

“…They implemented ontology reasoning and similarity measure but didn't consider relation between concepts. Shikha Agarwal, Arachana Singhal and Punam Pedi (2012) [10] used weighted concept frequency-inverse document frequency (cf-idf) with background knowledge of domain ontology, for classification of RSS news feed items. They have shown that a rich and comprehensive ontology can be successfully used as text classifier.…”

Section: Related Workmentioning

confidence: 99%

Dynamic Classification for Web Documents Using Semantic Knowledge (DBpedia)

Elkafrawy

Eldemerdash

2018

The Egyptian Journal of Language Engineering

View full text Add to dashboard Cite

we present a dynamic Web document Classification using semantic knowledge (DBpedia). We present a method for a dynamic Web document Classification and automatic classification. The proposed approach required only a domain ontology and a set of user predefined categories. Currently, most approaches to text classification represent document as (bag of words) and training the large set of documents to train the classifier. Our approach doesn't require a training set of documents. In our proposed method, we use DBpedia ontology as the main classifier, representing documents as (bag of concepts). We extract the terms from the document, extract their resources from DBpedia Spotlight, use Sparqle query to determine class ontology and map them to their concepts then we determine the best category.

show abstract

“…We have regularly downloaded news items from different RSS feeds and classify them semantically [2] into different categories using our designed news domain ontology as knowledge base [1]. News will be recommended to the authentic registered users on our designed portal.…”

Section: Our Approachmentioning

confidence: 99%

“…Ontology based classification of RSS feed news items have been done to give precise results based on user's requirements [2] Entities in news items have been identified using named entity recognition tool.…”

Section: Introductionmentioning

confidence: 99%

Proactive predictions to handle issues in recommendations

Agarwal

Singhal

Bedi

et al. 2014

2014 IEEE International Advance Computing Conference (IACC)

Self Cite

View full text Add to dashboard Cite

In today's online world users are suffering with the problem of information overload. To handle this problem, recommender systems assist users in giving required information by filtering out irrelevant information. So, most of the recommender systems mainly strive to achieve only accuracy in recommendations but this is not just what users want. Users require more coverage and diversity in recommendations mainly in the case of news domain which is highly dynamic in nature. To handle the issues of coverage and diversity we have worked on proactive predictions of those user interests which could not have been predicted by just user behavior analysis. User interest has been expanded on the basis of Concepts, sub concepts, entities, properties and relationships stored in our designed news domain ontology. Ontology design is based on news industry standards and careful study of the domain. It is also semantically annotated with context sensitive knowledge, extracted from external knowledge source DBpedia. It is obvious that only useful diversity in coverage will be appreciated by the users. Therefore along with improvement in coverage and diversity in recommendations, we have also experimentally evaluated the usefulness of diversity by implicit user behavior analysis. Improvements in results have been shown in the paper. It has also been noticed that user' profile contains interest score only in the news areas which user has ever touched. Due to which other news areas have no values for user interest causing sparse data problem. New score predicted for these unknown values in user's profile also helps to handle the issue of sparsity up to a certain extent. In our approach we have tried to handle this issue also along with above mentioned issues in news recommendation. Reduction in sparsity has been noticed in the users' profiles.Keywords-Ontology, semantic user profile, semantic linking, error free expansion of user profile, sparsity, coverage, diversity.

show abstract

Classification of RSS feed news items using ontology

Cited by 10 publications

References 13 publications

Linked Data Triples Enhance Document Relevance Classification

Linked Data Triples Enhance Document Relevance Classification

Dynamic Classification for Web Documents Using Semantic Knowledge (DBpedia)

Proactive predictions to handle issues in recommendations

Contact Info

Product

Resources

About