Decision Support System with K-Means Clustering Algorithm for Detecting the Optimal Store Location Based on Social Network Events

Nowadays, the business market is more complicated and comprises many challenges; it became more competitive and surrounded by high-risk patterns. Seeking for new technologies and adopting innovation is becoming an important and crucial issue to eliminate the complexity of the decision-making process and failure probability. Decision support system (DSS) is a computerized system that encompasses mathematical and analytical models, knowledge base and a user interface to help managers for making better decisions. This research aims to develop a decision support system based on K-means clustering algorithm to detect the optimal store location through social network events. Also, this research explains how to extract data from one social network channel "Instagram" using the "Octoparse API" as a web data extraction tool. K-means algorithm identifies k- number of centroids, and allocates every data point to the nearest cluster. As a result, we analyzed 12754 posts started on the 1st of January 2019. Cleaned data are transformed using Minimax and K-means algorithms. As an output, we have got json format data file with centres which are placed on the map to provide a better understanding. The Result of this research is a visualized map pointed with places to define the optimal location of a specific store at the selected region. The practical value of this DSS tool is to help users to make a more valuable and accurate decision which lead to a decrease in the probability of ineffective business decision and minimize the business losses.