Data mining Python or R small analysis
Budget: $10 – $30 USD
You will use data for 6 major cities.
To get started, download the data from the Course Task Data link in the moodle course page. Then remove only those for the city selected by your group. You can prepare the sample with both R and Python.
1. Preparation and study of data:
Does your data contain missing values and if so, how did you treat it?
Apply several R / Python aggregate statistics features to learn more about your data. Show pictures on the results screens and explain your findings.
Visualize the data with various relevant graphs and explain what they show.
2. Prediction:
Build a regression model with outcome variable review_scores_rating. Which independent variables did you include in the model? What is the RMSE of your model and what does it mean?
3. Classification:
Using the k-nearest neighbors algorithm, predict whether in your city, a property that accommodates 4 people, has 2 bathrooms and a "strict cancellation policy", would have a "cleaning fee".
Prepare a presentation of the results of your group's research, including a summary of your most important findings (executive summary).
You can use both R and Python (if a group knows and prefers it). The main goal is to show that you can put into practice some of the methods and techniques covered in the course.
To get started, download the data from the Course Task Data link in the moodle course page. Then remove only those for the city selected by your group. You can prepare the sample with both R and Python.
1. Preparation and study of data:
Does your data contain missing values and if so, how did you treat it?
Apply several R / Python aggregate statistics features to learn more about your data. Show pictures on the results screens and explain your findings.
Visualize the data with various relevant graphs and explain what they show.
2. Prediction:
Build a regression model with outcome variable review_scores_rating. Which independent variables did you include in the model? What is the RMSE of your model and what does it mean?
3. Classification:
Using the k-nearest neighbors algorithm, predict whether in your city, a property that accommodates 4 people, has 2 bathrooms and a "strict cancellation policy", would have a "cleaning fee".
Prepare a presentation of the results of your group's research, including a summary of your most important findings (executive summary).
You can use both R and Python (if a group knows and prefers it). The main goal is to show that you can put into practice some of the methods and techniques covered in the course.