Find Ideal Location for Business in Bangladesh

The dataset has 21 columns that carry the features (questions) of 988 respondents. The efficiency of any machine learning model is heavily dependent on its raw initial dataset. For this, we had to be extra careful in gathering our information. We figured out that for our particular problem, we had to go forward with data that was not only authentic but also versatile enough to get the proper information from relevant sources. Hence we opted to build our dataset by dispatching a survey questionnaire among targeted audiences. Firstly, we built the questionnaire with inquiries that were made after keen observation. Studying the behavior from our intended audience, we came up with factual and informative queries that generated appropriate data. Our prime audience were those who were highly into buying fashion accessories and hence we had created a set of questionnaires that emphasized on questions related to that field. We had a total of twenty one well revised questions that gave us an overview of all answers that were going to be needed within the proximity of our system. As such, we had the opportunity to gather over half a thousand authentic leads and concluded upon our initial raw dataset accordingly.