Further Data Cleaning Based on Business Context and Consideration #15
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This commit encompasses a series of data cleaning steps aimed at preparing the dataset for machine learning modeling. The following changes have been made: - Bounces and NewVisits Handling: Converted `NaN` values to `0` for the 'bounces' and 'newVisits' columns to indicate non-bounce sessions and returning visits, respectively, enhancing the binary feature's consistency. - TransactionRevenue Adjustment: Replaced `NaN` values with `0` in the 'transactionRevenue' column to represent sessions without any transactional revenue, addressing the sparsity of transactional data. - Keyword Column Removal: Dropped the 'keyword' column due to its high volume of `NaN` values and limited predictive value, streamlining the feature set. - IsTrueDirect Normalization: Converted `NaN` values to `False` in the 'isTrueDirect' column, clarifying the interpretation of direct versus indirect session access. - ReferralPath Column Removal: Dropped the 'referralPath' column due to its complexity and the high percentage of missing values, simplifying the dataset. - Campaign and Medium Enhancement: Updated the 'campaign' column to replace "(not set)" with "No Campaign" for clarity. Grouped less common mediums into an "other" category in the 'medium' column for model simplicity and removed rows where the medium was "(not set)". - Pageviews NaN Removal: Eliminated rows with `NaN` values in the 'pageviews' column, ensuring completeness in pageview data. These steps were taken to improve the dataset's quality and readiness for machine learning analysis, ensuring that each feature contributes meaningfully to the model's predictive accuracy.
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
After Abdul's initial cleaning, the dimensions of our dataset were: (903653, 28).
After further cleaning, I found additional values that were not usable for us (i.e., "not set", "not available in demo dataset"). I also considered that for our use cases, we would need entries with comprehensive geographical data, so I conducted a thorough inspection of geographical data for missing values (that could not be imputed by deduction). Finally, the dimensions of our dataset are now: (193578, 35).
The output of this code is the file called squeaky_clean_train_data.csv which can be used directly for future exploratory data analysis and modelling. This data set could also serve as the common point before the data is used for both use cases. This can only be confirm when Angel (Business Analyst for Use Case 1) approves this request.