Merge pull request #687 from why-aditi/main

Google Analytics Capstone Project

Google Analytics Capstone Project/Dataset/GCapstone.csv

@@ -0,0 +1,12 @@
+Interest in Large Language Model by Region throughout 2023,,,,,,Sourced from Google Trends Searches,,,,,Data is created by cumulating interest throughout periods of a month,
+,,,,,,,,,,,"If data includes numbers from other months, will be included in other months instead. (i.e. if data runs from November 26th to December 4th, the data will be counted for December)",
+Month:,January,February,March,April,May,June,July,August,September,October,November,December
+Regions:,,,,,,,,,,,,
+WorldWide,0,0,0,27,51,158,149,260,312,280,303,386
+China,0,7,0,46,35,73,161,303,214,217,277,234
+Singapore,6,0,5,33,69,120,218,205,324,237,320,328
+South Korea,3,0,0,17,25,30,153,163,252,238,313,307
+Japan,1,0,1,10,44,123,239,242,338,326,351,366
+United States,0,0,0,28,65,181,193,242,310,279,294,369
+,,,,,,,,,,,,
+"Max interest a month is four hundred, because each month has a period of four.",,,,,,,,,,,,

Google Analytics Capstone Project/Dataset/README.md

@@ -0,0 +1 @@
+dataset link: https://www.kaggle.com/datasets/fredericxiong/google-analytics-capstone-project

Google Analytics Capstone Project/Model/README.md

@@ -0,0 +1,59 @@
+## **PROJECT TITLE**
+
+### 🎯 **Goal**
+
+Create an analysis model for the Google analytics using machine learning.
+
+### 🧵 **Dataset**
+
+https://www.kaggle.com/datasets/fredericxiong/google-analytics-capstone-project
+
+### 🧾 **Description**
+
+Analysis of Interest in Generative AI across different Regions
+
+### 🧮 **What I had done!**
+
+Data Collection and Preparation -> EDA -> Model Training -> Model Validation -> Comparing the performance metrics of various models
+
+### 🚀 **Models Implemented**
+
+1. SARIMA
+2. ARIMA
+3. Linear Regression
+4. Random Forest
+5. LSTM
+
+### 📚 **Libraries Needed**
+
+1. NumPy
+2. Pandas
+3. Matplotlib
+4. Sci-kit learn
+
+### 📊 **Exploratory Data Analysis Results**
+
+<img src="https://github.com/why-aditi/ML-Crate/blob/main/Google%20Analytics%20Capstone%20Project/Images/download%20(1).png">
+<img src="https://github.com/why-aditi/ML-Crate/blob/main/Google%20Analytics%20Capstone%20Project/Images/download%20(2).png">
+<img src="https://github.com/why-aditi/ML-Crate/blob/main/Google%20Analytics%20Capstone%20Project/Images/download%20(3).png">
+<img src="https://github.com/why-aditi/ML-Crate/blob/main/Google%20Analytics%20Capstone%20Project/Images/download%20(4).png">
+<img src="https://github.com/why-aditi/ML-Crate/blob/main/Google%20Analytics%20Capstone%20Project/Images/download%20(5).png">
+<img src="https://github.com/why-aditi/ML-Crate/blob/main/Google%20Analytics%20Capstone%20Project/Images/download%20(6).png">
+
+### 📈 **Performance of the Models based on the Accuracy Scores**
+
+Mean Squared Error was used as performance metric
+1. SARIMA: 3025.1666666666665
+2. ARIMA: 0.015828689092572328
+3. Linear Regression: 0.15681863010490982
+4. Random Forest: 0.02206226453506559
+5. LSTM: 27538.882124875207
+
+
+### 📢 **Conclusion**
+
+ARIMA has turned out to be the best model with MSE 0.016.
+
+### ✒️ **Your Signature**
+
+Aditi Kala

Google Analytics Capstone Project/requirements.txt

@@ -0,0 +1,4 @@
+1. NumPy
+2. Pandas
+3. Tensorflow
+4. Sci-kit learn