Artificial intelligence and machine learning integration.

I’m excited and happy to share that I recently gained great exploratory of AI model -Data Science & model predictive analysis and implementation after completed my two recent AI predictive models projects successfully as part of Advanced Machine Learning from my Post Graduate Program in Artificial Intelligence and Machine Learning: Business Applications at UT Austin McCombs from #GreatLearning #AIML ?#ArtificalIntelligence ?#UTMCCombs ?#DataScience ?#EDA ?#MachineLearning. Happy Learning! #MyGreatLearning

AI and machine learning are not buzzwords anymore; they’re integral components of modern software development, setting new standards for functionality and performance. From predictive algorithms to automated code reviews, AI/ML technologies are enhancing efficiency and capabilities across various industries. AI/ML are the foundational and common building blocks for Gen AI LLMs moving to specialized subsets starting from

Advanced ML-> Deep Learning-> Neural Networks-> Gen AI.

"AI Models are nothing but a PURE mathematical problem" to help with logical reasoning, optimization and quantifying uncertainty using simple and powerful Statistical and probability analysis to predict the behavior. We classify this further in to

a) Supervised Learning (Solving common Regression & Classification Problems using labeled data sets)

b) UnSupervised Learning( Clustering technique, Dimensionality reduction unlabeled data sets, PCA)

c) Reinforced Learning(Punishment/reward-Robotics, Games).

Here is a quick overview of my recent project completion and sharing the knowledge of how this is executed using Powerful Python scripts using Anaconda/Google Colab using various libraries of Numpy, Pandas, matplotlib, seaborn, Sklearn.

?A) Course Project: Credit Card Users Churn Prediction

Bank saw a steep decline in the number of users of their credit card. Customers’ leaving credit card services would lead the bank to loss, so the bank wants to analyze the data of customers and identify the customers who will leave their credit card services and the reason for same – so that the bank could improve upon those areas.

B) Objective

1. Explore and visualize the dataset.
2. Build various classification models to predict if the customer is going to churn or not.
3. Identify the best possible model that will give optimized performance.
4. Optimize the model using appropriate techniques.
5. Generate a set of insights and recommendations that will help the bank.

C) AI/ML Approach solving the above problem.

Analyze the existing data of customers, bank and come up with a predictive model which is a Classification model to determine if a customer will leave the credit card services offered by the Bank or not and the reason behind it. Also, what needs to be done for the bank to improve their services so that customers do not renounce their credit cards.

Building five models using decision trees, Random Forest, bagging, and boosting methods.

D) Skills & Tools covered:

Powerful Python scripts using Anaconda/Google Colab with EDA, RANDOM FOREST, Decision Tree, Bagging, Boosting, SMOTE, Cross Validation, Data Preprocessing, Hyperparameter Tuning & using various libraries of Numpy, Pandas, matplotlib, seaborn, Sklearn.

E) Project Execution

?? Exploratory Data Analysis and Insights

?? Data pre-processing

?? Model Building - Original Data

?? Model building - Oversampled data

?? Model building - Undersampled data

?? Model Performance Improvement using Hyperparameter Tuning

?? Choose models that might perform better after tuning (tune at least 3 models out of 15 models built in the previous steps) - Provide proper reasoning for tuning that model - Tune the best 3 models obtained above using randomized search and metric of interest - Check the performance of 3 tuned models.

?? Model Performances and Final Model Selection

?? Actionable Insights & Recommendations to BANK.

?? Presentation / Notebook - Overall Code Quality.

Happy Learning!

?F) Project Evaluation Assessment-Grading result- Full 60/60.

G) Some of the quick snapshots of the assessment from model output:

<Axes: xlabel='Attrition_Flag', ylabel='Total_Amt_Chng_Q4_Q1'>        

Axes: xlabel='Income_Category', ylabel='Total_Amt_Chng_Q4_Q1'>        

Happy AI ML Learning! #MyGreatLearning

#AIML ?#ArtificalIntelligence ?#UTMCCombs ?#DataScience ?#EDA ?#MachineLearning#TeamWork?#EDA?#RANDOMFOREST?#Bagging?#Boosting?#SMOTE?#CrossValidation?#DataPreprocessing?#HyperparameterTuning?#MyGreatLearning

?

?

?

?

?