Comparison of a logistic regression and Naïve Bayes classifier in landslide susceptibility assessments: The influence of models complexity and training dataset size

Book is an important medium for teaching in higher education. It is facilitated by a library or a reading room which enabled student and teacher to fulfill their references for teaching and learning activities. For easy searching, each book classified by categories. In our institution, Information Technology Major of State Polytechnic of Malang, those categories are specifics to computer science topics. Every book entry need to be classified accordingly and to perform such task, one need to understand major keywords of the book title to correctly classify the books. The problem is, not all the librarian have such knowledge. Therefore manually classifying hundreds and even thousands of book is an exhausting work. This research is focused on automatic book classification based on its title using Naive Bayes Classifier and Log Probabilistic. The Log Probabilistic implementation is to solve the probability calculation result that is too small that cannot be represented in a computer programming floating points variable type. The algorithm then implemented in a web application using PHP and MySQL database. Evaluation has been done using Holdout method for 240 training dataset and 80 testing dataset resulting in 75% of accuration. We also tested the accuracy using K-fold Cross Validation resulting in 66.25% of accuration.

Download Full-text

Twitter Sentiment Analysis using Machine Learning Techniques

International Journal of Engineering and Advanced Technology - Regular Issue ◽

10.35940/ijeat.c6281.029320 ◽

2020 ◽

Vol 9 (3) ◽

pp. 4205-4209

Keyword(s):

Logistic Regression ◽

Sentiment Analysis ◽

Naive Bayes ◽

Naïve Bayes ◽

Machine Learning Techniques ◽

Naive Bayes Classifier ◽

Bayes Classifier ◽

Naïve Bayes Classifier ◽

Learning Techniques ◽

Social Media Platforms

Nowadays people share their views and opinions in twitter and other social media platforms, the way of recognizing sentiments and speculation in tweets is Twitter Sentiment Analysis. Determining the contradiction or sentiment of the tweets and then listing them into positive, negative and neutral tweets is the main classifying step in this process. The issue related to sentiment analysis is the naming of the correct congruous sentiment classifier algorithm to list the tweets. The foundation classifier techniques like Logistic regression, Naive Bayes classifier, Random Forest and SVMs are normally used. In this paper, the Naïve Bayes classifier and Logistic Regression has been used to perform sentiment analysis and classify based on the better accuracy of catagorizing Technique. The outcome shows that Naive Bayes classifier works better for this approach. Data pre-processing and feature extraction is realized as a portion of task.

Download Full-text

Application of Trust in Recommender Systems—Utilizing Naive Bayes Classifier

Computation ◽

10.3390/computation10010006 ◽

2022 ◽

Vol 10 (1) ◽

pp. 6

Author(s):

Korab Rrmoku ◽

Besnik Selimi ◽

Lule Ahmedi

Keyword(s):

Recommender Systems ◽

Prediction Accuracy ◽

Naive Bayes ◽

Real Life ◽

Naïve Bayes ◽

Training Dataset ◽

Naive Bayes Classifier ◽

Bayes Classifier ◽

Naïve Bayes Classifier ◽

Response Vector

Receiving a recommendation for a certain item or a place to visit is now a common experience. However, the issue of trustworthiness regarding the recommended items/places remains one of the main concerns. In this paper, we present an implementation of the Naive Bayes classifier, one of the most powerful classes of Machine Learning and Artificial Intelligence algorithms in existence, to improve the accuracy of the recommendation and raise the trustworthiness confidence of the users and items within a network. Our approach is proven as a feasible one, since it reached the prediction accuracy of 89%, with a confidence of approximately 0.89, when applied to an online dataset of a social network. Naive Bayes algorithms, in general, are widely used on recommender systems because they are fast and easy to implement. However, the requirement for predictors to be independent remains a challenge due to the fact that in real-life scenarios, the predictors are usually dependent. As such, in our approach we used a larger training dataset; hence, the response vector has a higher selection quantity, thus empowering a higher determining accuracy.

Download Full-text