I am currently in the process of choosing the data, trying to choose based on my desired outcomes. The last time, though, I considered choosing the economic route, but there is a drawback: there are fewer data points. I checked the “approved building permits” in this dataset today in an effort to verify that construction projects adhered to the minimal safety requirements set out by the State Building Code in order to protect everyone’s health and safety.

The dataset provides information on food inspections that take place in Boston according to pertinent sanitary rules and standards. At least once a year, businesses that serve food are examined, and high-risk facilities get follow-up inspections. I can infer from this data how well the Boston area’s food chain functions and whether the big brands truly are that huge or whether they function similarly to any other food chain. I’ll use this data to forecast the future of the food industry, and I’ll try to use visualization to develop other visualization libraries other than Matplot and Altair.

Later on, I do more analysis on the qualities of the available data will determine whether time series analysis or traditional predictive modeling is used. Time series approaches are the most effective means of deciphering temporal sequences’ complexities. They provide a tailored approach to seeing and predicting long-term trends that generic models might overlook. When navigating the data landscape, it is helpful to be aware of the advantages and disadvantages of each data navigation strategy so that we can choose the right tool for the job.

In my next blog, I will try to figure out the topic in which I sorted 3 to topics from the Analyze Boston data hub and will start working on the model.