Open AccessPrediction of internet user satisfaction levels in Bangladesh using data mining and analysis of influential factors
Author(s) -
Md. Hasan Imam Bijoy,
Sumiya Alam Akhi,
Md. Ali Ashraf Nayeem,
Md. Mahbubur Rahman,
Md. Jueal Mia
Publication year - 2022
Publication title -
bulletin of electrical engineering and informatics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.251
H-Index - 12
ISSN - 2302-9285
DOI - 10.11591/eei.v11i2.3617
Subject(s) - the internet , internet access , user satisfaction , internet users , reliability (semiconductor) , context (archaeology) , computer science , world wide web , geography , power (physics) , physics , archaeology , human–computer interaction , quantum mechanics
Today the world has already acknowledged as a global village by the inter-net which has technologically evolved into a significant performance instrument for individuals, businesses, and countries seeking to achieve betterment. This study is based on data mining techniques to predict the satisfaction level of internet users from the context of Bangladesh. After conducting a public survey with 18 questions, we were able to acquire 451 responses from participants. Data for user satisfaction was associated with end-user characteristics including certain getting high speed, internet packages, cable type of Wi-Fi connection with targeting various age groups and occupations. The research's most key conceptual breakthrough was the reliability of magnitude predictions of user satisfaction level based on their experience with internet use. The empirical findings indicate that people in Bangladesh have high expectations in existing internet technology, and they are very dissatisfied with their facilities of internet use and to measure satisfaction level related with monthly limit of the Wi-Fi packages and the elements affecting internet speed. Several classifier models were applied to our dataset and among them, Random Forest (RF) performance reaches the top position with 91.53% accuracy.