Machine Learning Algorithms for Depression Detection and Their Comparison

Textual emotional intelligence is playing a ubiquitously important role in leveraging human emotions on social media platforms. Social media platforms are privileged with emotional content and are leveraged for various purposes like opinion mining, emotion mining, and sentiment analysis. This data analysis is also levered for the prevention of online bullying, suicide prevention, and depression detection among social media users. In this article, we have designed an automatic depression detection of online social media users by analyzing their social media behavior. The designed depression detection classification can be effectively used to mine user's social media interactions and one can determine whether a social media user is suffering from depression or not. The underlying classifier is made using state-of-art technology in emotional artificial intelligence which includes LSTM (Long Short Term Memory) and other machine learning classifiers. The highest accuracy of the classifier is around 70% of LSTM and for SVM the highest accuracy is 81.79%. We trained the classifier on the datasets that are widely used in literature for emotion mining tasks. A confusion matrix of results is also given.

Online Fake Review Detection Using Supervised Machine Learning And BERT Model

Online shopping stores have grown steadily over the past few years. Due to the massive growth of these businesses, the detection of fake reviews has attracted attention. Fake reviews are seriously trying to mislead customers and thereby undermine the honesty and authenticity of online shopping environments. So far, various fake review classifiers have been proposed that take into account the actual content of the review. To improve the accuracies of existing fake review classification or detection approaches, we propose to use BERT (Bidirectional Encoder Representation from Transformers) model to extract word embeddings from texts (i.e. reviews). Word embeddings are obtained in various basic methods such as SVM (Support vector machine), Random Forests, Naive Bayes, and others. The confusion matrix method was also taken into account to evaluate and graphically represent the results. The results indicate that the SVM classifiers outperform the others in terms of accuracy and f1-score with an accuracy of 87.81%, which is 7.6% higher than the classifier used in the previous study [5].