ICT702 Data Analysis Project For the Dataset Cleansing

Dataset cleansing and preparing

 For this data analysis project, there are three datasets are provided which includes the following, Region.csv, StoreDetails.csv and WeeklySales.csv.  First the WeeklySales dataset contains the information about the Store ID, Date and Weekly_Sales amount. StoreDetails.csv includes Store, Date, Temperature, Unemployment rate in the regions and finally the Region.csv file includes the store numbers in different regions.  

In order to get the details of the sales form the different regions, sales and different other factors we merged the dataset depending the common column store in the different datasets.

Moreover, the merged dataset is also checked by the pandas package in order to remove the null values as well as inconsistency in the newly created dataset.

Insights from the dataset through the plots

In order to find the rate of unemployment and its relation with the sales of ice-cream in different regions we merged the datasets using the pandas package of python.  The overall statistics of the merged dataset is provided in the following table,

For the merged dataset, there is total 6435 rows of data in it.  For the weakly sales of the maximum mean and minimum value are provided by 172225.55, 19213.485 and 711.110.  in case of temperature the maximum and minimum temperature in all the regions recorded are 100.14 and -2.060. at last the maximum, minimum and mean rate of unemployment (among all the regions) according to the dataset is provided by, 14.31,3.87 and 7.999.   

From the weekly sales data sets, we found the following trend about the sales according to the different regions present in the data set  

From the above plot it can be derived that the Region B has the maximum amount of the sales.  The C region is at second position from the sales point of view. Following is plot that depicts the rate of un-employability in the different regions

Following chart shows the un-employability rates in different regions  

By comparing the plots for the sales of the ice-cream in the different regions and the un-employability in different regions it can be said that for the region B, the sales of ice cream high with the unemployment rate. On the other hand, in case   of region E, the sales of ice cream are lowest compared to the highest rate of unemployment.  

Now to measure the performance of different stores in the different regions we plotted the sales of the ice cream grouping by the store numbers in the data sets. 

From the above image, it can be stated that, complete merged dataset consists of total 45 stores in different regions.  Accumulated maximum sales is done by the store number 13 whereas the lowest sales are by the store number 36.

Distribution of un-employability in the dataset is given by the following histogram plot; 

From the above histogram plot, it can be said that the histogram is skewed right as there are only few values at the increasing   value of the un employability. Thus it can be said that, the most general value for the un-employability in the regions are between 6 to 10.

Conclusion

Sales data analytics is considered as an important internal business function which is more than just presenting figures and sales numbers to the management in the organization.  With much more in-depth methodologies in recording and crunching data the data about sales and other factors (having impact on the sales) can be presented in an easily-digestible format for a non-technical person too. Through the above analysis, the organization should try to improve its business in the regions (described in the sales analysis) by using proper strategy.