SWATAC: A Sentiment Analyzer using One-Vs-Rest Logistic Regression

Abstract

This paper describes SWATAC, a system built for SemEval-2015's Task 10 Subtask B, namely the Message Polarity Classification Task. Given a tweet, the system classifies the sentiment as either positive, negative, or neutral. Several preprocessing tasks such as negation detection, spell checking, and tokenization are performed to enhance lexical information. The features are then augmented with external sentiment lexicons. Classification is done with Logistic Regression using a one-vsrest configuration. For the test runs, the system was trained using only the provided training tweets. The classifier was successful, with an F1 score of 58.43 on the official 2015 test data, and an F1 score of 66.64 on the Twitter 2014 progress data.

Thyroid Syndrome Detection using Machine Learning Algorithms: A Comparative Analysis

“…For algorithm implementation, we employed the scikit-learn library and adopted the one-vs-rest (OvR) training scheme [29]. To handle class imbalances, we assigned weights inversely proportional to the frequencies of each class in the input data.…”

Evaluation of classification ability of logistic regression model on SERS data of miRNAs

“…"Multinomial" mode was generally considered to be more accurate in multiple classification cases, but our results demonstrated that when all the hyper-parameters were set to the same except the pattern ("ovr" or "multinomial"), this conclusion was not tenable. "Ovr" classification method regards the classification task as multiple binary logistic regressions, while "multinomial" mode is to choose two classes from all data to estimate, and then, to choose another two classes from the rest data to judge till the end [29]. However, in the actual processing, only by calculating of the specific data could demonstrate which parameter configuration is suitable for this situation most.…”