Comparative Assessment of Extractive Summarization: TextRank, TF-IDF and LDA

Rani, Ujjwal; Bidhan, Karambir

doi:10.37398/jsr.2021.650140

Cited by 21 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use Newspaper3k library to retrieve the summary of the text. It uses TextRank algorithm [ 27 ] to rank the sentences in the article and selects top k sentences as the summary. TextRank is an extractive text summarisation technique which generates a summary based on the sentences presented in the input text itself [ 28 ].…”

Section: Methodsmentioning

confidence: 99%

Topical analysis of migration coverage during lockdown in India by mainstream print media

Agarwal

Sarkar

2022

PLoS ONE

View full text Add to dashboard Cite

Implementing countrywide lockdown measures in India, from March 2020 to May 2020 was a major step to deal with the COVID -19 pandemic crisis. The decision of country lockdown adversely affected the urban migrant population, and a large section of them was compelled to move out of the urban areas to their native places. The reverse migration garnered widespread media attention and coverage in electronic as well as print media. The present study focuses on the coverage of the issue by print media using descriptive natural language text mining. The study uses topic modelling, clustering, and sentiment analysis to examine the articles on migration issues during the lockdown period published in two leading English newspapers in India- The Times of India and The Hindu. The sentiment analysis results indicate that the majority of articles have neutral sentiment while very few articles show high negative or positive polarity. Descriptive topic modelling results show that transport, food security, special services, and employment with migration and migrants are the majorly covered topics after employing Bag of Words and TF-IDF models. Clustering is performed to group the article titles based on similar traits using agglomerative hierarchical clustering.

show abstract

Section: Methodsmentioning

confidence: 99%

Topical analysis of migration coverage during lockdown in India by mainstream print media

Agarwal

Sarkar

2022

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Since users’ interest in an item happens to be reflected in the text information, we incorporate TF-IDF (Rani and Bidhan, 2021), a common statistical method to describe the importance of a keyword in a document or corpus into the user’s interest degree model. We assess the user’s interest degree in an attribute by computing the proportion of the user’s historical ratings on the attribute in that on all attributes, weighted by the logarithm of the inverse frequency of the attribute of the items in the dataset.…”

Section: The Architecture Of the Uicf Recommendation Frameworkmentioning

confidence: 99%

UICF: a new user-item composite filtering recommendation framework by leveraging temporal semantics

Wei,

Liu,

Xian

et al. 2024

IJICC

View full text Add to dashboard Cite

PurposeThe collaborative filtering algorithm is a classical and widely used approach in product recommendation systems. However, the existing algorithms rely mostly on common ratings of items and do not consider temporal information about items or user interests. To solve this problem, this study proposes a new user-item composite filtering (UICF) recommendation framework by leveraging temporal semantics.Design/methodology/approachThe UICF framework fully utilizes the time information of item ratings for measuring the similarity of items and takes into account the short-term and long-term interest decay for computing users’ latest interest degrees. For an item to be probably recommended to a user, the interest degrees of the user on all the historically rated items are weighted by their similarities with the item to be recommended and then added up to predict the recommendation degree.FindingsComprehensive experiments on the MovieLens and KuaiRec datasets for user movie recommendation were conducted to evaluate the performance of the proposed UICF framework. Experimental results show that the UICF outperformed three well-known recommendation algorithms Item-Based Collaborative Filtering (IBCF), User-Based Collaborative Filtering (UBCF) and User-Popularity Composite Filtering (UPCF) in the root mean square error (RMSE), mean absolute error (MAE) and F1 metrics, especially yielding an average decrease of 11.9% in MAE.Originality/valueA UICF recommendation framework is proposed that combines a time-aware item similarity model and a time-wise user interest degree model. It overcomes the limitations of common rating items and utilizes temporal information in item ratings and user interests effectively, resulting in more accurate and personalized recommendations.

show abstract

“…Based on the principle of semantic retrieval, we use a vector space model based on TF/IDF feature term weights [23] to implement a feature vector representation of natural language sequences. The initial frequency statistics of sequence word frequencies sf ij in the affective computing system are performed using the following formula.…”

Section: Vector Space Representation Modelmentioning

confidence: 99%

Research on Semisupervised Affective Interaction Technology in Immersive Virtual Teaching Environment

Tang

Zhang

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

The separation of time and space in immersive virtual teaching makes students unable to realize emotional communication, which may affect students’ mental health. In recent years, the use of affective computing technology to solve the problem of affective loss in distance education has become a key research topic. In order to realize the problem of emotion interaction in immersive virtual teaching, a semisupervised support vector machine- (SVM-) based affective interaction model was proposed. First, the natural language sequences of students in the virtual teaching environment are preprocessed using a statistical-based framing method, and mutual information and expected cross-entropy are used as feature selection methods. Then, a vector space model based on TF/IDF feature term weights is proposed to implement the feature vector representation of natural language sequences. Finally, after the constructed sentiment space, a semisupervised SVM is employed as the classifier to complete the affective interaction computation. The experimental results of emotion classification show that the proposed model is able to determine and understand the emotional state more accurately than other traditional models and significantly improves the training speed. In addition, the proposed model can provide emotional encouragement or emotional compensation according to the specific emotional state of the learner.

show abstract

Comparative Assessment of Extractive Summarization: TextRank, TF-IDF and LDA

Cited by 21 publications

References 16 publications

Topical analysis of migration coverage during lockdown in India by mainstream print media

Topical analysis of migration coverage during lockdown in India by mainstream print media

UICF: a new user-item composite filtering recommendation framework by leveraging temporal semantics

Research on Semisupervised Affective Interaction Technology in Immersive Virtual Teaching Environment

Contact Info

Product

Resources

About