Theory finder is an intuitive application designed to redefine academic research for Information Systems professionals. By using classification & similarity algorithms, application swiftly navigates database which consisted of information from IS Theories website and keywords about theories. It also offers users the flexibility to specify the number of theories they wish to explore, granting them precise control over their research process. Aim of the project is to deliver highly relevant theories in response to user queries. Seamlessly bridging user inputs with an extensive collection of theories, this platform ensures researchers effortlessly access the relevant theories for their research.

Our Dash-based data visualization project revolves around empowering users with an intuitive tool for dissecting intricate financial loan approval default data sourced from Kaggle. The tool offers a systematic approach in three phases: initially presenting seven individual variables through graphs and statistical insights, providing users a foundational understanding of each variable's significance. Next, it allows users to select multiple graphs, affording them the freedom to compare any two variables of their choice, fostering a personalized exploration of correlations. Furthermore, the tool dives deeper into the dataset by displaying graphs involving three variables, enabling users to unravel complex relationships and gain a more comprehensive understanding of the factors influencing loan approvals and defaults. Through its interactive nature and reliance on Dash's capabilities, this visualization project not only simplifies data comprehension but also engages users in a dynamic exploration, making it an invaluable resource for researchers, analysts, and individuals keen on deciphering the complexities of financial loan approval dynamics.

Designed a Decision Tree algorithm using Python, harnessing the readily accessible Iris dataset—a built-in resource in Python's libraries. Nested within the architecture of this algorithm, my objective entailed sculpting a predictive model that harnesses four pivotal parameters: precisely, the intricate dimensions encapsulating the floral length and width. The ultimate ambition revolved around precisely determining the authentic label linked with every individual flower, consequently revealing its unadulterated botanical categorization. As a final touch, I conducted a comprehensive evaluation of the model's performance and visualized the outcome of this evaluation.

Developed a multiple linear regression algorithm in Python using the diabetes dataset, an inherent part of Python's built-in datasets. This dataset encompasses information from individuals dealing with diabetes, featuring essential attributes like BMI, age, blood pressure, and glucose levels. These attributes play a pivotal role in prognosticating the progression of diabetes in patients. Within this algorithmic framework, my focus was on constructing a robust multiple linear regression model. This model delves into the intricate connections between independent variables and the dependent variable. To gauge the effectiveness of the model, I conducted a comprehensive evaluation, including an in-depth analysis of variance table.