•  
  •  
 

Abstract

People's need for food is increasing day after day due to the large population increase that has occurred in human society, which has led to a significant increase in the activity of food product manufacturing companies in the world. Hence the necessary need to evaluate the performance of food production companies. We found that the best way to evaluate the performance of these companies is through social media comments, because it is the newest and most common method among people. In this paper, 4 machine learning techniques were relied upon to evaluate participants' comments in evaluating food production companies after using a lexicon of Arabic language terms in general and the Iraqi dialect in particular. These terms relate to positive and negative words and expressions. Then training through machine learning algorithms. Random Forest (RF), Naive Bayes (NB), Rough Set Theory (RST), and Support Vector Machine (SVM) algorithms were tested. Experiments have shown that RST is superior to the other three. RST achieved (96.13%), SVM achieved (95.75%), although RF achieved (94.1%) and NB achieved (87.1%).

Keywords

Assessment model, Facebook comments, Food product, Iraqi dialect machine learning

Subject Area

Computer Science

Article Type

Article

First Page

1366

Last Page

1378

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Download

Share

COinS