Smart Computing and Systems Engineering - 2022 (SCSE 2022)

Permanent URI for this collectionhttp://repository.kln.ac.lk/handle/123456789/25392

Browse

Now showing 1 - 3 of 3

Machine Learning Approach to Predict Mental Distress of IT Workforce in Remote
(Department of Industrial Management, Faculty of Science, University of Kelaniya Sri Lanka, 2022) Gamage, Sanduni Nilushika; Asanka, P. P. G. Dinesh
When considering online workers, due to the emergence of the coronavirus pandemic prevailing in the world, employees have been restricted to work remotely for a prolonged period. All the working arrangements are now based at home than before. Since this has been novel to society, the impact caused by this crisis on people is unknown in the short or long term. Since various factors can cause mental distress among online workers, periodic screening for mental distresses such as anxiety, depression, and stress is necessary for health and well-being. The causes of mental distress are multifactorial. They include socio-demographic, biological, economic, environmental, occupational, and psychological aspects. This paper proposes a concept of a screening system to predict mental distress given the external features associated with individuals, using supervised machine learning approaches and identifying the employees prone to higher risk and referring them early to professional assistance. The study was conducted concerning the circumstances in a pandemic era considering COVID-19 as the case study. The study was done with remote IT workers in Sri Lanka who work as a part of a software development team. 481 professionals participated in the study and were selected based on selection criteria and appropriate encoding techniques were utilized to encode categorical variables where most important 25 features were detected among 60 features using feature selection. Finally, classification techniques such as Random Forest, SVM, XGBoost, CatBoost, decision tree, and Naïve Bayes were used for modeling by which the CatBoost algorithm in overall measures outperformed other algorithms with a predictive accuracy of 97.1%, precision of 97.4%, recall of 99.7%, and f1 measure is 98.5%.
Personalized Classification of Non-Spam Emails Using Machine Learning Techniques
(Department of Industrial Management, Faculty of Science, University of Kelaniya Sri Lanka, 2022) Dinendra, Harsha; Rajapakse, Chathura; Asanka, P. P. G. Dinesh
With the advent of computer networks and communications, emails have become one of the most widely accepted communication means, which is faster, more reliable, cheaper, and accessible from anywhere. Due to the increased use of email communications, day-to-day computer users; particularly corporate users, find it cumbersome to filter the most important and urgent emails out of the large number of emails they receive on a given business day. Enterprise email systems are able to automatically identify spam emails but still, there are many non-urgent and unimportant emails among such non-spam emails which cannot be filtered by conventional spam filter programs. Though it may be feasible to set up some static rules and categorize some of the e-mails, the practicality and sustainability of such rules are questionable due to the magnitude of such rules, and the validity period as such rules may become redundant after some time. Thus, it is desired to have an email filtering system for non-spam emails to filter unimportant emails, based on the user’s past behaviour. Despite the availability of research on identifying spam e-mails in the area of further classifying the non-spam e-mails, is lacking. The purpose of this research is to provide a machine learning-based solution to classify non-spam e-mails considering the importance of such e-mails. As part of the research, several machine learning models have been developed and trained using non-spam e-mails, based on the personal mailbox of the first author of this research. The results showed a significant accuracy, particularly with a decision tree, random forests and deep neural network algorithms. This paper presents the modelling details and the results obtained accordingly.
Sentiment Analysis of ASOS Product Reviews Using Machine Learning Algorithms by Comparing Several Models
(Department of Industrial Management, Faculty of Science, University of Kelaniya Sri Lanka, 2022) Soundearajah, Sahithya; Asanka, P. P. G. Dinesh
Digital ratings are crucial in improving international customer communications and impacting consumer purchasing trends. To obtain important data from a massive number of customer reviews, they must be sorted into positive and negative opinions. Sentiment analysis is a computational method for extracting emotive information from a text. In this particular research, over 3000 reviews have been obtained from the ASOS website and classified into three different sentiments: excellent, average, and bad. The obtained reviews have been pre-processed, then feature extraction is applied to the pre-processed data to remove the redundant data. Finally, distinct machine learning algorithms will be utilized to build disparate models. This research is vital as it allows the ASOS organization to gain insight into how consumers perceive about specific issues and detect urgent issues such as delivery delays and misplaced packages in the current time period before the issue goes outof control. The key results of this research show that the Nu- Support Vector Classification model obtained the highest accuracy score of 85.99% and the lowest accuracy score of 51.47% was obtained for the AdaBoost classifier model.