Extracting the needs of the labor market in Riyadh through Twitter using text classification techniques

Main Article Content

najlaa Alsadan

Abstract

Recently, Twitter has attracted a great deal of spread and attention. It is one of the most common social networking sites for sharing ideas, chats, and transfer of information and news through text. The labor market is where the supply and demand for jobs meet, with employees satisfying employer needs for certain services. On the one hand, certain jobs are being eliminated, while others are being replaced by new jobs that were not even possible a few years ago. In this paper, we focused on labor market classification of twitter data belonging to Riyadh city and written in Modern Standard Arabic. We want to classify Arabic jobs’ tweets to determine the trending of required job in Riyadh city. Twitter’s API was used to collect tweets related to labor market. Five different classifiers were used on the dataset namely; Support Vector Machine (SVM), Multinomial Naive Bayes (M-NB), Decision Tree (DT), Gradient Boosting Classifier (GBC), and Random Forest (RF).to classify the tweets based on their related job classes. We evaluated our work by four different measures which are Precision, Recall, Accuracy and F-measure. We made a comparison between the five classifiers based on those measures. The results show that RF achieved the best Accuracy and F-measure, and it equals 93.62%, 93% respectively. In addition, we found that the trend of labor market needed was administrative jobs “وظائ٠ادارية. “.  The percentages of that job class that related to 384 jobs about 19.2%.

Downloads

Download data is not yet available.

Article Details

Section
Articles