Improving TF-IDF with Singular Value Decomposition (SVD) for Feature Extraction on Twitter

Kadhim, Ammar Ismael; Cheah, Yu-N; Hieder, Inaam Abbas; Ali, Rawaa

doi:10.23918/iec2017.16

Cited by 7 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tokenization describes interpreting and grouping isolated tokens to generate higher-level tokens in addition to dividing strings into fundamental processing units. Word tokenization includes preprocessed raw texts are divided into textual units 18,19 . During the dataset cleaning stage, the columns from the datasets that weren't needed for processing were removed.…”

Section: Data Pre-processingmentioning

confidence: 99%

Fake News Detection Model Basing on Machine Learning Algorithms

Taha,

Jabar,

Mohammed

2024

Baghdad Sci.J

View full text Add to dashboard Cite

The rapid growth of the internet and easy communication has made it quick and simple to create and spread news. Social media users now generate and share more information than before, but some of it is false and unrelated to reality. Detecting false information in text is challenging, even for experts who need to consider multiple factors to determine authenticity. Malicious misinformation on social media negatively affects societies, especially during crises like terrorist attacks, riots, and natural disasters. To minimize the harmful impact, it is crucial to identify rumors quickly. This study aims to build a learning model for detecting fake news. This research paper relies on finding and analyzing the characteristics of the text, then the words are converted into features using TF-IDF technology, after that the highest-ranking features are identified for the purpose of studying and distinguishing the spread of news, whether it is real or fake using machine learning techniques. Finally, the Logistic Regression, Decision Tree, Gradient Boosting and Random Forest algorithm has been adapted. The accuracy of Logistic Regression is 0.985, Random Forest (0.989) whereas the accuracy of Decision Tree is 0.994 and Gradient Boosting (0.9949), respectively.

show abstract

Section: Data Pre-processingmentioning

confidence: 99%

Fake News Detection Model Basing on Machine Learning Algorithms

Taha,

Jabar,

Mohammed

2024

Baghdad Sci.J

View full text Add to dashboard Cite

show abstract

“…As shown in [18], dimension reduction to TF-IDF can be achieved using Singular Value Decomposition (SVD) which is a dimensionality reduction method that works well with sparse matrices. Equation 4 shows how SVD is calculated.…”

Section: Feature Extractionmentioning

confidence: 99%

COVID-19 Hakkındaki Türkçe Tweetlerde LSTM Ağı Kullanılarak Duygu Sınıflandırması

Cataltas

Baykan

2023

Konya Journal of Engineering Sciences

View full text Add to dashboard Cite

As Covid-19 pandemic affected everyone in various aspects, people have been expressing their opinions on these aspects mostly on social media platforms because of the pandemic. These opinions play a crucial role in understanding the sentiments towards the pandemic. In this study, Turkish tweets on Covid-19 topic were collected from March 2020 to January 2021 and labelled as positive, negative, or neutral in terms of sentiment using BERT which is a pre-trained text classifier model. Using this labelled dataset, a set of experiments were carried out with SVM, Naive Bayes, K-Nearest Neighbors, and CNN-LSTM model machine learning algorithms for binary and multi-class classification tasks. Results of these experiments have shown that CNN-LSTM model outperforms other machine learning algorithms which are used in this study in both binary classification and multi-class classification tasks.

show abstract

“…This diagonal matrix has the square root of eigenvalues of V or U. The SVD offers a strong mathematical foundation for the field of text analytics [3]. In SVD, the matrix A represents the text document as a high dimension vector space model.…”

Section: A Singular Value Decompositionmentioning

confidence: 99%

Feature Selection Methods for Mining Social Media

Mageshwari*,

Aroquiaraj

2019

IJITEE

View full text Add to dashboard Cite

People can share their thoughts and opinion through Social Media which can easily widespread. So many public issues and political views are also discussed on social media. HIV/AIDS is also one of the important topics discussed. This work aims to classify HIV/AIDS related twitter data. Since the twitter data is highly dimensional, it is essential to do reduce dimensionality of the data to attain better classification results. Tweets are collected using keyword search and necessary preprocessing steps are carried out. Then feature extraction methods such as Bag of Words (BOW) model and TF-IDF are implemented. Singular Value Decomposition (SVD) and Principal Component Analysis (PCA) techniques are used for dimensionality reduction. Finally, classification is carried out and the results are discussed.

show abstract

Improving TF-IDF with Singular Value Decomposition (SVD) for Feature Extraction on Twitter

Cited by 7 publications

References 9 publications

Fake News Detection Model Basing on Machine Learning Algorithms

Fake News Detection Model Basing on Machine Learning Algorithms

COVID-19 Hakkındaki Türkçe Tweetlerde LSTM Ağı Kullanılarak Duygu Sınıflandırması

Feature Selection Methods for Mining Social Media

Contact Info

Product

Resources

About