Abstract:
There is a great potential for the growth of e-commerce in Bangladesh due to the growth
in internet connectivity and usage. Knowledge and solving the issues related to
consumer concern are necessary to maximize the benefit from this sector. Thus, this
research explores customer sentiment and issues concerning e-commerce businesses in
Bangladesh by employing a Google Form-based survey with 599 entries and 16
features. Demographic information, web use history and target characteristic such as
‘Good’ and ‘Bad’ satisfaction with the service are incorporated into the dataset. For the
given problem to predict consumer perception, machine learning models such as GB,
RF, BNB, SVC, DT were applied. It was fulfilled by the Random-Forest classifier with
high accuracy of 99.16%, surpassing other algorithms. This speaks to the model’s
ability to detect subtleties in terms of time-series data and forecast consumer
satisfaction accurately under the conditions of e-commerce in Bangladesh. The gaps
linked to several constituents of consumer satisfaction, including the product quality,
delivery efficacy, and the methods of payment, were also established via the analysis.
Essential plan points to consider when seeking to boost the position of e-commerce
ventures are; logistics and delivery, sustainability, and ethical factors. The rights of
consumers and fair competition policies are promoted by encouraging policymakers to
put into place relevant regulations. This present research adds knowledge to the subject
of e-commerce development in emerging economies such as Bangladesh, thus
providing information that may be useful in designing the proper policies that may help
steady development of this business model and improving consumer confidence in
electronic commerce transactions.
