HIB Las Vegas Strip Dataset Analysis Write Up Essay

Task:

Develop a predictive model and analysis write up to predict how many times a review was deemed helpful by other users. (Helpful votes).

Dataset:

LasVegasTripAdvisorReviews-1.xlsx Download LasVegasTripAdvisorReviews-1.xlsx

Dataset Source

http://archive.ics.uci.edu/ml/datasets/Las+Vegas+Strip (Links to an external site.)

Moro, S., Rita, P., & Coelho, J. (2017). Stripping customers’ feedback on hotels through data mining: The case of Las Vegas Strip. Tourism Management Perspectives, 23, 41-52.

Deliverables:

1. Analysis Write Up

Save this as a .pdf file under the following naming convention: FIRSTNAME_LASTNAME_A1_Write_Up.pdf

• Present your best TWO insights (maximum 100 words per insight)
• Make one actionable recommendation based on your analysis and offer recommendations for business implementation (maximum 200 words)
• State your final model’s highest R-Square value, rounded to three decimal places

2. Data Analysis and Code

Save this as a Jupyter Notebook (.ipynb) file under the following naming convention: FIRSTNAME_LASTNAME_A1_Analysis.ipynb

Tell the story of your analysis through:

• exploratory data analysis
• feature treatment and engineering
• utilizing appropriate modeling techniques

3. Final Model

Save this as a Python (.py) script under the following naming convention: FIRSTNAME_LASTNAME_A1_Model.py

Model will be assessed on:

• R-Square value on unseen data (randomly seeded)
• Processing speed (see coding requirements below)
• Appropriateness for the problem at hand

• Being submitted as a .py script (Note: coding files submitted as anything other than a .py script will receive a one-letter grade deduction)

Coding Requirements

Both the analysis code and the final model must meet the following requirements:

• runs from start to finish in under one minute (based on Prof. Chase’s computer)
• be coded in Python or R (R may only be used if it runs within your Python script)
• be well commented
• avoid “data dumping” (i.e. avoid any unnecessary output or graphics)
• run without errors or bugs