This data science project aims to build a predictive model to estimate the sale price of houses in Boston based on various factors such as number of rooms, crime rate, accessibility to highways, etc. The project will use the Boston Housing dataset that contains 506 observations with 13 attributes, collected by the US Census Bureau in 1978. The project will involve data cleaning, exploratory data analysis, feature engineering, and building a regression model to predict the prices of houses. The model performance will be evaluated using metrics such as mean squared error (MSE) and R-squared. Built a client facing API using flask

This end-to-end Data Science project is designed to predict the test scores of students based on the number of hours they studied. It considers several other variables that may affect a student's academic performance, including Gender, Ethnicity, Parental level of education, Lunch, and Test preparation course.

The first step in this project is data cleaning, which involves cleaning and preprocessing the data to remove any inconsistencies, missing values, and errors. Once the data is cleaned, the next step is exploratory data analysis, where we visualize and analyze the data to gain insights into the relationships between variables and to identify any patterns or trends.

After exploratory data analysis, we move on to feature engineering, which involves selecting and transforming the most important variables in the dataset to enhance the predictive power of the model. Once we have selected the relevant features, we then split the data into training and testing sets and proceed with model building.

In this project, we may use various regression models such as Linear Regression, Random Forest Regression, or Gradient Boosting Regression, to train the data and create the best predictive model. After building the model, we evaluate its performance using various metrics such as Mean Squared Error, Root Mean Squared Error, Mean Absolute Error, and R-Squared.

Finally, we interpret the results and draw conclusions about the factors that most significantly impact student test scores. Overall, this project aims to provide a comprehensive end-to-end process for predicting student test scores using Data Science techniques and methods.

Predicting Swiggy Ratings: A Data-Driven Journey

Imagine. You order your favorite biryani on Swiggy, eagerly anticipating its arrival. But will it be a 'Delightful' experience, or a 'Disappointing' disaster? Predicting delivery satisfaction is crucial for Swiggy, and that's where my project comes in.

---

The Challenge: Develop a machine learning model to forecast 'Rating Category' for Swiggy food orders, leveraging 17 diverse features like delivery time, price, and cuisine type.

---

My Approach:
Data Wrangling: Diving into the depths of the dataset, I tackled missing values, transformed categorical variables, and identified unique identifiers for removal. This cleansed data was my canvas for further exploration.

Base Models: Before diving deep, I established baselines with Logistic Regression, Decision Trees, Random Forests, Naive Bayes, and XGBoost. Decision Trees and Random Forests emerged as early contenders, with Random Forest shining for its balanced performance.

Class Imbalance: The data leaned towards positive reviews. To ensure fairness, I employed oversampling techniques, creating a more balanced playing field for my models.

Exploratory Data Analysis (EDA): I donned my detective hat, uncovering outliers and skewness within the data. Outliers received special attention, while skewness was corrected to ensure my models wouldn't be swayed by imbalanced distributions.

Model Refinement: With a cleansed and balanced dataset, I revisited my models. The results were impressive! Both Random Forest and XGBoost soared, showcasing high accuracy, precision, recall, and F1-scores for both rating categories.

Feature Selection: Time to declutter! Using Recursive Feature Elimination, I identified the most influential features, crafting a leaner and meaner model. While performance remained strong, this step provided valuable insights into feature importance.

---

The Verdict:
Random Forest and XGBoost emerged as the top performers, capable of accurately predicting Swiggy's 'Rating Category' with balanced precision and recall. This opens doors for exciting possibilities! Swiggy can leverage these insights to:

---

Beyond the code:
This project wasn't just about lines of algorithms. It was a journey of understanding data, uncovering hidden patterns, and ultimately, using insights to drive real-world impact. In a world fueled by online deliveries, predicting satisfaction has become a game-changer, and I'm thrilled to have contributed to this exciting frontier.

Predicting Concrete Compressive Strength

Remember the classic Tic-Tac-Toe? This project takes it to a whole new level by injecting the magic of artificial intelligence! Built for a college submission, it combines the nostalgia of the game with the power of Python, HTML, and JavaScript to create a truly unbeatable opponent.

Imagine a Tic-Tac-Toe board where you can actually challenge the computer and always end up in a nail-biting draw. That's exactly what this project delivers! The secret weapon? The Min-Max algorithm, a clever strategy used by AI to predict future moves and make the best possible decision.

Under the hood, Python handles the game's logic and ensures the AI's razor-sharp moves. Meanwhile, HTML and JavaScript work together to craft a sleek and user-friendly interface. You'll see the familiar X's and O's dance across the screen as you make your mark, while the computer responds with uncanny precision.

But it's not just about winning (or rather, not losing)! This project also explores the fascinating world of AI and how it can enhance our entertainment experiences. From the code structure to the testing process, there's a treasure trove of learning opportunities hidden within this game.

So, are you ready to face off against an AI mastermind in a classic battle of wits? Dive into this project and discover the power of combining technology and fun! You might just find yourself learning a thing or two about artificial intelligence along the way.

Putting Actitime on Autopilot: Streamlining Tasks with Selenium

Imagine having a virtual assistant that can effortlessly handle repetitive tasks on the Actitime website—that's exactly what this automation project achieves! It taps into the power of Selenium, Java, and Eclipse to create a trusty sidekick that works tirelessly behind the scenes, saving you time and effort.

---

Here's how it works:
Actitime, Meet Selenium: Selenium takes the reins, acting as a skilled web browser controller. It can navigate through pages, click buttons, fill in forms, and even make decisions—just like a human user!

Java Powers the Brain: Java, a versatile programming language, provides the logic and structure for the automation scripts. It's the mastermind behind the scenes, telling Selenium what to do and when.

Eclipse Keeps Things Organized: Eclipse, a popular development environment, offers a user-friendly workspace to create, manage, and run the automation tests. It's like a neat office for your project, keeping everything in its place.

---

Key Tasks Automated:

Task 001: Effortlessly creates new customers on Actitime, handling all the clicks and form entries.

Task 002: Smoothly adds new tasks, guiding Selenium through selecting customers, projects, and filling in task details—a breeze!

Task 003: Repeats the task creation process with precision, ensuring accuracy and consistency.

The purpose of this project is to make a java based application which includes different OOPS concepts such as inheritance, encapsulation, polymorphism and abstraction etc.

I have used java concept like static and non-static members ,interface ,getter-setter method and scanner class

I have used Scanner class to take input from user and switch case to display output

Developed a fully functional calculator application with a user-friendly graphical interface using Python and the Kivy framework.Employed Kivy's efficient layout and widget classes (BoxLayout, GridLayout, Label, Button) to construct the calculator's visual elements.Established clear event handling for button presses to capture user input and execute calculations.Integrated Python's built-in eval() function for robust expression evaluation.Implemented error handling to gracefully manage potential syntax errors and provide informative feedback to the user.Incorporated a clear button to enable users to reset the calculator for subsequent calculations.

This project utilizes UiPath to streamline the organization of music files within a specified directory. By leveraging the TagLib library, the automation extracts album metadata from each audio file, allowing for intelligent categorization. The workflow systematically processes each file, cleans the album names to eliminate any unnecessary details, and creates dedicated folders for each album.

Once the album folders are established, the corresponding music files are moved into their respective directories based on the cleaned album names. This project not only improves the accessibility and management of large music collections but also demonstrates the effectiveness of automation in enhancing file organization processes. The end result is a structured and easily navigable music library that saves time and effort for users.

This project utilizes UiPath to streamline the organization of music files within a specified directory. By leveraging the TagLib library, the automation extracts album metadata from each audio file, allowing for intelligent categorization. The workflow systematically processes each file, cleans the album names to eliminate any unnecessary details, and creates dedicated folders for each album.

Once the album folders are established, the corresponding music files are moved into their respective directories based on the cleaned album names. This project not only improves the accessibility and management of large music collections but also demonstrates the effectiveness of automation in enhancing file organization processes. The end result is a structured and easily navigable music library that saves time and effort for users.

This UiPath project automates the process of extracting shared links from WhatsApp Web group chats. Users interact with a UiPath Form that allows them to either select all available groups or choose specific ones using a multi-select dropdown. The bot opens WhatsApp Web, searches for each selected group, navigates to the 'Media, Links and Docs' section, and scrapes all available links from the 'Links' tab. It compares each link against an existing Excel file to ensure only new links are recorded, reducing duplication and maintaining clean data.

Each newly found link is saved in the Excel file along with the corresponding group name for easy tracking. The project uses browser automation, data scraping, and Excel activities.