Data Science // Evaluation Parameters

This is an evaluation guideline/documentation for data science submissions, where candidate must have worked on functional prototype, and would have visualisations, data points and possible code to back the analsis/findings.

Sample Splitup/Breakdown:

• Code : 15
• Graphs/Visualization/Maps : 10
• Usability : 10

Code

• Functionality and efficiency of the code and a choosen method (i.e. Python, R, Tableau, SAS, Microsoft BI, etc.)
• If the solution covers valid ternds, data points and/or statistical analysis on given problem/dataset
• Relevance of data points and produced output w.r.t. data analysis, processing, probability and statistics

Graphs/Visualization/Charts/Maps

• Use of sane and relevant Graph/Chart (Pareto/Bar, Pie/Circle, Histogram, Dot Plot, Scatter, Line, Cosmographs, Area, Waterfall, Radar, Bubble, etc)
• Elaborative self-explanatory design
• Use of Maps/Visualization (Comparisons, Proportions, Relationships, Hierarchy, Distribution, Movement/Flow, Pattern, Geographical, Vector, Heat, etc) whereever required

Usability (Findings/Impact/Validation)

• How creative is the finding from the analytical point of view; and is it what we've asked for, is it more than what we've asked for; and other supporting deliverables submitted by the candidate

• Validation of user submitted findings with existing answers set by problem setter/data scientist

• Use of cloud providers, available resources and/or delivering more than what has been asked

• Extra efforts/Uniqueness/creativity in given submission(s)

• Sample solution: data-science-sample-answer

— Note: This is raw guideline, and can be altered/customized as per the requirement and mutual discussion between Client and Evaluation Team.