Maximizing Sales Through Strategic Customer Insights
Maximizing Sales Through Strategic Customer Insights
Embedded Files
In a collaborative effort with three classmates, I contributed to increasing home insurance sales for Regent Insurance Company. Our task involved analyzing current customers and predicting potential future clients. This project greatly enhanced my teamwork skills, as well as my ability to handle technical aspects.
In a collaborative effort with three classmates, I contributed to increasing home insurance sales for Regent Insurance Company. Our task involved analyzing current customers and predicting potential future clients. This project greatly enhanced my teamwork skills, as well as my ability to handle technical aspects.
Additionally, I led the creation of a compelling model presentation, resulting in our team receiving an honorable mention in the class competition.
Additionally, I led the creation of a compelling model presentation, resulting in our team receiving an honorable mention in the class competition.
My Contribution:
My Contribution:
In spearheading the technical aspects of our Case Competition Project, I embarked on a rigorous journey of data exploration and analysis. Employing various advanced techniques, I meticulously honed in on optimizing feature selection to enhance model efficacy. Through iterative experimentation and refinement, I strategically identified and removed features devoid of variance, ensuring a streamlined dataset primed for predictive modeling. Leveraging the power of LASSO regression, I further pruned our feature set, fostering model interpretability and performance. Subsequently, recognizing the challenge posed by our imbalanced data, I engineered a solution by implementing an XGBOOST model, tailored to mitigate bias and amplify predictive accuracy. My strategic approach and technical expertise were instrumental in empowering our team to deliver impactful results and emerge as frontrunners in the competition.
In spearheading the technical aspects of our Case Competition Project, I embarked on a rigorous journey of data exploration and analysis. Employing various advanced techniques, I meticulously honed in on optimizing feature selection to enhance model efficacy. Through iterative experimentation and refinement, I strategically identified and removed features devoid of variance, ensuring a streamlined dataset primed for predictive modeling. Leveraging the power of LASSO regression, I further pruned our feature set, fostering model interpretability and performance. Subsequently, recognizing the challenge posed by our imbalanced data, I engineered a solution by implementing an XGBOOST model, tailored to mitigate bias and amplify predictive accuracy. My strategic approach and technical expertise were instrumental in empowering our team to deliver impactful results and emerge as frontrunners in the competition.
The following presentation summarizes the problem statement, the data analysis steps, the model development, and the model interpretability analysis using SHAP values.
Presentation slide Group 7 for presentation (Group 10)FINAL (1).pdfThe code is available in the following GitHub repository in two parts:
1- Exploratory data analysis:
https://github.com/israajashaami/Codes/blob/main/EDA.R
2- Final XG-BOOST model:
https://github.com/israajashaami/Codes/blob/main/final%20model.R
Page updated
Google Sites
Report abuse