Reference no: EM132354383 , Length: word count:1400
Predictive Analytics Assignment -
Case study - Olist Store is the largest department store in the Brazilian marketplaces. Olist connects several small businesses in Brazil to customers so those Brazilian merchants can sell their products to their customers through Olist. Shipment of products to customers is done using Olist logistics partners. Customers buy/order products through Olist store and sellers get notified to then fulfil the order. Customers will then receive the product/s they have ordered (or are notified of an estimated delivery date). A satisfaction survey is sent to the customer by email too so they can write some feedback comments on the purchase experience.
The data set that you are given contains a large number of customer orders and customer satisfaction review scores for each order/purchase experience. You are asked to explore and analyse this data set and then, create insights as to what patterns relate to the different customer satisfaction levels (1 to 5, with 5 being most satisfied). You are expected to analyse the data set using exploratory methods, develop appropriate predictive models to test your exploratory hypotheses, then evaluate and further improve your models.
The original data set contained 100K product orders from 2016 to 2018; however, to avoid computational power issues, a sample of 20K orders have been extracted for this assignment. It is expected that no further sampling will be done on this data set. It is strongly recommended that you read the data definitions to better understand the data context and terminology before you attempt the development of the analytic solution/s.
Assignment submission - The submission template file will need to be used for the final submission of your report. It is essential that the executive summary section is aimed at a non-technical reader and the remaining parts of the report target a data/business analyst.
Your final deliverables must include: i) the final report according to the submission template, and ii) all RapidMiner files (in the RMP or XML format). It is recommended that partial work be submitted regularly - as you go. You may include tables, charts, or screenshots of your analysis and models in your report. The consistency of your RapidMiner file(s) will be checked against the results in your report. You should not modify the data file provided for this assignment before importing it into RapidMiner. The use of Excel or any other analytic tool is therefore not permitted.
Assessment - Your work will be assessed based on the following criteria:
RapidMiner is to be used for all assignment tasks.
The advanced tasks will need to be attempted and reported only after meeting the expectations (no marks will be given for advanced work where basic solutions have not been attempted or reported).
The assessment rubric given in the assignment folder should be carefully read and consulted.
Attachment:- Predictive Analytics Assignment File.rar