Submission: Group 8: Predict and Classify Online Shopper Intention

Submitting authors: Calvin Choi, Nour Abdelfattah, Sai Pusuluri, Sana Shams

Repository: https://github.com/DSCI-310-2024/DSCI-310_Group-Project_Group8_Purchasing-Intent-Analysis/releases/tag/v2.0.0

Abstract/executive summary:

Given the surge of online shopping, online retailers may get a lot of site traffic but what ultimately matters is whether or not users finalize their purchase. Marketing and User Experience teams are tasked with optimizing a site’s interface and content in order to improve customer retention and the site’s revenue. Given this, understanding customer browsing behaviour and web page features is crucial for not only improving the user’s experience, but also maximizing the retailer’s revenue.

This project aims to analyze various features of online shopper’s sessions on a site to predict whether the customer makes a purchase. We will use the dataset, Online Shoppers Purchasing Intention dataset from the UCI Machine Learning Repository.

Editor: @ttimbers

Reviewer: Yunxuan Zhang, Dia Zavery, Hanlin Zhao, Olivia Lam

[ ] I agree to abide by DSCI 310's Code of Conduct during the review process.

Data analysis review checklist

Reviewer:

Conflict of interest

[x] As the reviewer I confirm that I have no conflicts of interest for me to review this work.

Code of Conduct

[x] I confirm that I read and will adhere to the MDS code of conduct.

General checks

[x] Repository: Is the source code for this data analysis available? Is the repository well organized and easy to navigate?
[x] License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?

Documentation

[x] Installation instructions: Is there a clearly stated list of dependencies?
[x] Example usage: Do the authors include examples of how to use the software to reproduce the data analysis?
[x] Functionality documentation: Is the core functionality of the data analysis software documented to a satisfactory level?
[x] Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Code quality

[x] Readability: Are scripts, functions, objects, etc., well named? Is it relatively easy to understand the code?
[x] Style guidelides: Does the code adhere to well known language style guides?
[x] Modularity: Is the code suitably abstracted into scripts and functions?
[x] Tests: Are there automated tests or manual steps described so that the function of the software can be verified? Are they of sufficient quality to ensure software robsutness?

Reproducibility

[x] Data: Is the raw data archived somewhere? Is it accessible?
[x] Computational methods: Is all the source code required for the data analysis available?
[x] Conditions: Is there a record of the necessary conditions (software dependencies) needed to reproduce the analysis? Does there exist an easy way to obtain the computational environment needed to reproduce the analysis?
[x] Automation: Can someone other than the authors easily reproduce the entire data analysis?

Analysis report

[x] Authors: Does the report include a list of authors with their affiliations?
[x] What is the question: Do the authors clearly state the research question being asked?
[x] Importance: Do the authors clearly state the importance for this research question?
[x] Background: Do the authors provide sufficient background information so that readers can understand the report?
[x] Methods: Do the authors clearly describe and justify the methodology used in the data analysis? Do the authors communicate any assumptions or limitations of their methodologies?
[ ] Results: Do the authors clearly communicate their findings through writing, tables and figures?
[x] Conclusions: Are the conclusions presented by the authors correct?
[x] References: Do all archival references that should have a DOI list one (e.g., papers, datasets, software)?
[x] Writing quality: Is the writing of good quality, concise, engaging?

Estimated hours spent reviewing:

Review Comments:

Please provide more detailed feedback here on what was done particularly well, and what could be improved. It is especially important to elaborate on items that you were not able to check off in the list above.

You've put together an impressive analysis, with clear explanations of the models under consideration and a thorough discussion of the trade-offs involved. Your attention to detail in these areas is commendable. However, I noticed in the pdf document that there are placeholders for table and figure references—such as "Table ??" and "Figure ??." To maintain the quality and clarity of your report, it would be beneficial to ensure these references are correctly linked to the actual tables and figures they correspond to. The data section of your project has a lot of datasets which is really thorough. It might help to sort them into 'raw' and 'processed' folders for easier navigation. Also, it's not immediately clear why the raw data is split into features and targets. Adding a quick explanation for this could clear up any confusion. In your Readme file, there's a small discrepancy in the instructions for setting the kernel environment. It mentions ProjectMilestone1_env, but the correct environment name is project_env. Updating this detail will help avoid any confusion for those replicating the analysis. Overall, I enjoyed reading through the analysis, it made everything super clear. You guys did a great job on the project!

Attribution

This was derived from the JOSE review checklist and the ROpenSci review checklist.

Data analysis review checklist

Reviewer:

Conflict of interest

[x] As the reviewer I confirm that I have no conflicts of interest for me to review this work.

Code of Conduct

[x] I confirm that I read and will adhere to the MDS code of conduct.

General checks

[x] Repository: Is the source code for this data analysis available? Is the repository well organized and easy to navigate?
[x] License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?

Documentation

[x] Installation instructions: Is there a clearly stated list of dependencies?
[x] Example usage: Do the authors include examples of how to use the software to reproduce the data analysis?
[x] Functionality documentation: Is the core functionality of the data analysis software documented to a satisfactory level?
[x] Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Code quality

[x] Readability: Are scripts, functions, objects, etc., well named? Is it relatively easy to understand the code?
[x] Style guidelides: Does the code adhere to well known language style guides?
[x] Modularity: Is the code suitably abstracted into scripts and functions?
[x] Tests: Are there automated tests or manual steps described so that the function of the software can be verified? Are they of sufficient quality to ensure software robsutness?

Reproducibility

[x] Data: Is the raw data archived somewhere? Is it accessible?
[x] Computational methods: Is all the source code required for the data analysis available?
[x] Conditions: Is there a record of the necessary conditions (software dependencies) needed to reproduce the analysis? Does there exist an easy way to obtain the computational environment needed to reproduce the analysis?
[x] Automation: Can someone other than the authors easily reproduce the entire data analysis?

Analysis report

[x] Authors: Does the report include a list of authors with their affiliations?
[x] What is the question: Do the authors clearly state the research question being asked?
[x] Importance: Do the authors clearly state the importance for this research question?
[x] Background: Do the authors provide sufficient background information so that readers can understand the report?
[x] Methods: Do the authors clearly describe and justify the methodology used in the data analysis? Do the authors communicate any assumptions or limitations of their methodologies?
[x] Results: Do the authors clearly communicate their findings through writing, tables and figures?
[x] Conclusions: Are the conclusions presented by the authors correct?
[x] References: Do all archival references that should have a DOI list one (e.g., papers, datasets, software)?
[x] Writing quality: Is the writing of good quality, concise, engaging?

Estimated hours spent reviewing:

1.5

Review Comments:

Guys from Group 8, I think you've done a great job. Here are just some of my tips and comments.

Addressing Class Imbalance I noticed in your analysis that the class imbalance in the 'Revenue' feature was acknowledged, which is great! It's really important to understand how such imbalances can skew our model's performance, especially towards the majority class. I'm curious if you considered implementing any specific strategies to tackle this imbalance, such as SMOTE (Synthetic Minority Over-sampling Technique) or undersampling techniques? It might be beneficial to explore these options to enhance the model's ability to predict minority class instances more accurately.
Expanding Model Evaluation Metrics Your choice to use Precision as an evaluation metric is insightful, particularly for focusing on the model's performance in predicting positive instances. Given the class imbalance issue, incorporating additional metrics like Recall, F1 Score, and the AUC-ROC curve could provide a more rounded view of the model's effectiveness across different scenarios. These metrics can help highlight if the model disproportionately favors one class over another and could be a great addition to your analysis.
Detailed Preprocessing and Feature Engineering Steps I found the sections on data preprocessing and feature engineering really interesting! It's clear that a lot of thought went into preparing the data for modeling. However, I think it would be even more helpful if you could dive deeper into some of the choices made during this process. For instance, explaining why certain scaling methods were chosen or detailing the rationale behind dropping or transforming specific features could give readers a clearer understanding of your project's methodology. It could also offer insights into how different preprocessing steps impact the overall model performance.
For test file One small thing I noticed, though, is that it seems like the tests are missing comments or docstrings explaining what each one does. I totally get how tests might seem self-explanatory, especially when you're deep into writing them. But I found that adding a short comment or docstring can be super helpful, not just for others trying to understand what's being tested (like me 😅), but also for our future selves when we come back to this code after a while. It's sort of like leaving breadcrumbs for anyone who follows, making it easier to grasp the purpose and logic behind each test. Just a thought! Keep up the awesome work!

Attribution

This was derived from the JOSE review checklist and the ROpenSci review checklist.

Data analysis review checklist

Reviewer:

Conflict of interest

[x] As the reviewer I confirm that I have no conflicts of interest for me to review this work.

Code of Conduct

[x] I confirm that I read and will adhere to the MDS code of conduct.

General checks

[x] Repository: Is the source code for this data analysis available? Is the repository well organized and easy to navigate?
[x] License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?

Documentation

[x] Installation instructions: Is there a clearly stated list of dependencies?
[x] Example usage: Do the authors include examples of how to use the software to reproduce the data analysis?
[x] Functionality documentation: Is the core functionality of the data analysis software documented to a satisfactory level?
[x] Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Code quality

[x] Readability: Are scripts, functions, objects, etc., well named? Is it relatively easy to understand the code?
[x] Style guidelides: Does the code adhere to well known language style guides?
[x] Modularity: Is the code suitably abstracted into scripts and functions?
[x] Tests: Are there automated tests or manual steps described so that the function of the software can be verified? Are they of sufficient quality to ensure software robsutness?

Reproducibility

[x] Data: Is the raw data archived somewhere? Is it accessible?
[x] Computational methods: Is all the source code required for the data analysis available?
[x] Conditions: Is there a record of the necessary conditions (software dependencies) needed to reproduce the analysis? Does there exist an easy way to obtain the computational environment needed to reproduce the analysis?
[x] Automation: Can someone other than the authors easily reproduce the entire data analysis?

Analysis report

[x] Authors: Does the report include a list of authors with their affiliations?
[x] What is the question: Do the authors clearly state the research question being asked?
[x] Importance: Do the authors clearly state the importance for this research question?
[x] Background: Do the authors provide sufficient background information so that readers can understand the report?
[x] Methods: Do the authors clearly describe and justify the methodology used in the data analysis? Do the authors communicate any assumptions or limitations of their methodologies?
[x] Results: Do the authors clearly communicate their findings through writing, tables and figures?
[x] Conclusions: Are the conclusions presented by the authors correct?
[x] References: Do all archival references that should have a DOI list one (e.g., papers, datasets, software)?
[x] Writing quality: Is the writing of good quality, concise, engaging?

Estimated hours spent reviewing:

~ 1.5 hours

Review Comments:

Excellent job in terms of specificity with the usage instructions. Clearly explained all possible methods and provided step-by-step instructions on how to run the analysis, from creating the environment to generating the report. This ensures that users can execute the analysis successfully with minimal effort and without needing in-depth knowledge of the file structure of your project.
Further expanding the level of communication regarding each completed or resolved issue should be encouraged. I've noticed that 'sanash43' is excelling in this regard, but there's room for improvement among other team members. It's typical for challenges to emerge throughout the development process, often discussed across various communication channels like group chats or verbal exchanges. However, as the project scales up, it's easy to lose track of the specific problems tackled. Consequently, maintaining comprehensive records detailing the issues and their solutions becomes important for future reference. This ensures a clear understanding of the project's history and aids in avoiding the recurrence of previously addressed issues.
As an additional consideration, I think the analysis methods for this project are restricted to the contents of DSCI100. I've noticed that your analysis includes preprocessing, machine learning model selection, and metrics for model evaluation, which were topics covered in CPSC330. It would be nice to avoid employing these advanced techniques in this analysis, despite your excellent interpretation of the model results.
I've noted that Figures 2, 3, 4, 5, and 6 in your report might benefit from an increased plot size, as the current size makes it somewhat challenging to read the contents without zooming in. One alternative approach could be to vertically append the pie chart and bar chart within each figure, rather than presenting them side by side horizontally. This adjustment should improve the readability of the figures.

Attribution

This was derived from the JOSE review checklist and the ROpenSci review checklist.

Data analysis review checklist

Reviewer: ole-lam

Conflict of interest

[x] As the reviewer I confirm that I have no conflicts of interest for me to review this work.

Code of Conduct

[x] I confirm that I read and will adhere to the MDS code of conduct.

General checks

[x] Repository: Is the source code for this data analysis available? Is the repository well organized and easy to navigate?
[x] License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?

Documentation

[x] Installation instructions: Is there a clearly stated list of dependencies?
[x] Example usage: Do the authors include examples of how to use the software to reproduce the data analysis?
[x] Functionality documentation: Is the core functionality of the data analysis software documented to a satisfactory level?
[x] Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Code quality

[x] Readability: Are scripts, functions, objects, etc., well named? Is it relatively easy to understand the code?
[x] Style guidelides: Does the code adhere to well known language style guides?
[x] Modularity: Is the code suitably abstracted into scripts and functions?
[x] Tests: Are there automated tests or manual steps described so that the function of the software can be verified? Are they of sufficient quality to ensure software robsutness?

Reproducibility

[x] Data: Is the raw data archived somewhere? Is it accessible?
[x] Computational methods: Is all the source code required for the data analysis available?
[x] Conditions: Is there a record of the necessary conditions (software dependencies) needed to reproduce the analysis? Does there exist an easy way to obtain the computational environment needed to reproduce the analysis?
[x] Automation: Can someone other than the authors easily reproduce the entire data analysis?

Analysis report

[x] Authors: Does the report include a list of authors with their affiliations?
[x] What is the question: Do the authors clearly state the research question being asked?
[x] Importance: Do the authors clearly state the importance for this research question?
[x] Background: Do the authors provide sufficient background information so that readers can understand the report?
[x] Methods: Do the authors clearly describe and justify the methodology used in the data analysis? Do the authors communicate any assumptions or limitations of their methodologies?
[x] Results: Do the authors clearly communicate their findings through writing, tables and figures?
[x] Conclusions: Are the conclusions presented by the authors correct?
[x] References: Do all archival references that should have a DOI list one (e.g., papers, datasets, software)?
[x] Writing quality: Is the writing of good quality, concise, engaging?

Estimated hours spent reviewing: 1 hour 15 mins

Review Comments:

This was a really great analysis with a really applicable problem to explore! You guys were very through explaining each step of your analysis. I thought it was really smart to use a confusion matrix to support your conclusion as you had started your analysis under the knowledge that your data had a class imbalance.
I noticed in your pdf there are a few format errors, but very easy to fix! I noticed that your in-text citations didn't have the right syntax, your figures were referenced as ?? or [INSERT LATER], and your pie charts themselves could have been made more readable by perhaps using an alternative chart like a stacked bar chart or waffle chart.
I also noticed that in your preprocessing you immediately remove all NA values and duplicate values. Although I agree both of these could contribute to more class imbalance especially duplicates, I think it is important to explore these duplicate and NA rows before immediately dropping them in case you lose a substantial amount of data.
I also noticed that your pdf left out a lot of information from your initial analysis like recall and F1 score as well as feature engineering. Perhaps it was a time constraint thing (I totally understand), but I thought these aspects of your original analysis could have been further elaborated on and enhance your final submission. Especially if you explained perhaps why you thought page_values had such an impact on your analysis, as it felt like you added feature engineering just to do it, and not because you were necessarily making an argument for why it was necessary to incorporate it.

Overall such a cool project, I was really looking for little things to improve!

Attribution

This was derived from the JOSE review checklist and the ROpenSci review checklist.

DSCI-310-2024 / data-analysis-review-2024