Vectorization of Text Documents for Identifying Unifiable News Articles

Singh, Anita; Shashi, M.

doi:10.14569/ijacsa.2019.0100742

Cited by 85 publications

(32 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The process of converting textual content into meaningful numerical representation is known as vectorization [39]. Many feature engineering techniques exist in literature to vectorize texts, such as Bag-of-Words [40], Word2Vec [41], FastText [42], Glove [43] and Term Frequency-Inverse Document Frequency (TF-IDF) [44].…”

Section: ) Feature Extractionmentioning

confidence: 99%

Score and Rank Level Fusion Algorithms for Social Behavioral Biometrics

Tumpa

Gavrilova

2020

IEEE Access

View full text Add to dashboard Cite

The goal of a biometric system is to recognize individuals based on their unique physiological or behavioral traits. Online Social Networking (OSN) platforms have become an integral part of the daily life of individuals, where they leave a recognizable trail of behavioral information. Social Behavioral Biometric (SBB), being an emerging trend, focuses on such trails to distinguish between individuals. This research investigates the impact of users' writing profiles on OSN to conclude whether such profiles contribute to SBB. The distinctiveness of the SBB features that are extracted from the social behavioral data of Twitter is studied. A person identification system that relies on users' writing profiles, reply, retweet, shared weblink, trendy topic networks and temporal profiles is proposed. Score and rank level weighted fusion algorithm performance is compared on a social interaction database of 241 Twitter users. The experimental results establish that the users' writing profiles have the highest impact over other social biometric features and that score level fusion algorithms perform better than rank level fusion on SBB. The proposed system has achieved recognition rate of 99.45% at rank-1 after cross-validation using genetic algorithm based score level fusion algorithm. The system outperformed all prior researches on SBB in terms of identification accuracy.

show abstract

Section: ) Feature Extractionmentioning

confidence: 99%

Score and Rank Level Fusion Algorithms for Social Behavioral Biometrics

Tumpa

Gavrilova

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…They did not observe an advantage with the utilization of ELMo, compared to the commonly used, like GloVe or the random indexing approach. Singh et al [6] propose a vectorization approach based on word targets, to identify unifiable news articles. They define a framework for identifying news related to trending topics/hashtags.…”

Section: Literature Reviewmentioning

confidence: 99%

“…However, our approach uses non-labeled datasets: patents, journals and learning resources. In addition, it is the only one that uses learning resources and patents, and only another one uses a scientific publication dataset in its analysis [6]. In regard to the used techniques, our work is interested in feature extraction methods to transform text documents into a list of features that can be easily used and understood, like BM25 and TF-IDF, and methods of document vectorization to create numerical features using statistical analysis, like LDA, LSA, and Doc2Vec.…”

Section: Comparison With Previous Workmentioning

confidence: 99%

“…Finally, most of previous papers used supervised metrics in order to test the quality of the methods, in contrast with our work where a different approach is presented using several types of unsupervised metrics: one based on information theory (entropy) and the others based on document similarity. Only [6] considers unsupervised metrics, and there are several works that consider document similarity, but none information theory metrics. Furthermore, in the context of digital educational contents, there are not many works.…”

Section: Comparison With Previous Workmentioning

confidence: 99%

See 1 more Smart Citation

Comparison and Evaluation of Different Methods for the Feature Extraction from Educational Contents

et al. 2020

View full text Add to dashboard Cite

This paper analyses the capabilities of different techniques to build a semantic representation of educational digital resources. Educational digital resources are modeled using the Learning Object Metadata (LOM) standard, and these semantic representations can be obtained from different LOM fields, like the title, description, among others, in order to extract the features/characteristics from the digital resources. The feature extraction methods used in this paper are the Best Matching 25 (BM25), the Latent Semantic Analysis (LSA), Doc2Vec, and the Latent Dirichlet allocation (LDA). The utilization of the features/descriptors generated by them are tested in three types of educational digital resources (scientific publications, learning objects, patents), a paraphrase corpus and two use cases: in an information retrieval context and in an educational recommendation system. For this analysis are used unsupervised metrics to determine the feature quality proposed by each one, which are two similarity functions and the entropy. In addition, the paper presents tests of the techniques for the classification of paraphrases. The experiments show that according to the type of content and metric, the performance of the feature extraction methods is very different; in some cases are better than the others, and in other cases is the inverse.

show abstract

“…Text Vectorization: In this step, the text data is converted into numerical form [14]. Word Embedding algorithm is a process in which the input text is converted to the equivalent number representation.…”

mentioning

confidence: 99%

Automated Identification of Disaster News for Crisis Management using Machine Learning and Natural Language Processing

Domala

Dogra

Masrani

et al. 2020

2020 International Conference on Electronics and Sustainable Communication Systems (ICESC)

View full text Add to dashboard Cite

We are living in unprecedented times and anyone in this world could be impacted by natural disasters in some way or the other. Life is unpredictable and what is to come is unforeseeable. Nobody knows what the very next moment will hold, maybe it could be a disastrous one too. The past cannot be changed but it can act constructively towards the betterment of the current situation, 'Precaution is better than cure'. To be above this uncertain dilemma of life and death situations, 'Automated Identification of Disaster News for Crisis Management is proposed using Machine Learning and Natural Language Processing'. A software solution that can help disaster management websites to dynamically show the disaster relevant news which can be shared to other social media handles through their sites. The objective here is to automatically scrape news from English news websites and identify disaster relevant news using natural language processing techniques and machine learning concepts, which can further be dynamically displayed on the crisis management websites. The complete model is automated and requires no manual labor at all. The architecture is based on Machine Learning principles that classifies news scraped from top news websites using a spider-scraper into two categories, one being disaster relevant news and other being disaster irrelevant news and eventually displaying the relevant disaster news on the crisis management website.

show abstract

Vectorization of Text Documents for Identifying Unifiable News Articles

Cited by 85 publications

References 7 publications

Score and Rank Level Fusion Algorithms for Social Behavioral Biometrics

Score and Rank Level Fusion Algorithms for Social Behavioral Biometrics

Comparison and Evaluation of Different Methods for the Feature Extraction from Educational Contents

Automated Identification of Disaster News for Crisis Management using Machine Learning and Natural Language Processing

Contact Info

Product

Resources

About