Vimalraaj Sathiapal

As a highly motivated MSc Data Science graduate, I am seeking to apply my skills and knowledge to a challenging role in the field of Data Science. My academic background has equipped me with a vast skill set, including proficiency in programming languages like Python, R and SQL, strong analytical and problem-solving skills and the ability to think critically, enabling me to effectively tackle complex data challenges. I have also developed a solid foundation in Statistical Analysis, Mathematical Modelling and Data Manipulation. In addition to my technical abilities, I am a proactive team player with excellent communication skills . I thrive in a collaborative environment and am looking forward to work enthusiastically in a dynamic team.

IMDB Movie Data Analysis

• Conducted an extensive analysis on the IMDB dataset to gain insights into movie trends and audience preferences.
• Utilized data preprocessing techniques to clean and transform the dataset, ensuring data integrity and consistency.
• Performed Exploratory Data Analysis (EDA) to identify patterns, correlations and outliers, enabling deeper understanding of the dataset. Applied statistical methods to analyse variables such as movie genres, IMDB score, budget etc.
• Presented findings and insights through visually appealing data visualisations, including pie charts and bar graphs.

Physical Activity Monitoring Data Analysis

• Conducted a comprehensive analysis on the PAMAP2 Dataset, a dataset widely for physical activity monitoring and human motion recognition.
• Utilised signal processing techniques, such as feature extraction and time series analysis, to extract meaningful insights from accelerometer and gyroscope data.
• Conducted in-depth EDA to understand patterns and relationships between physical activity features and classified the activities into light, moderate and vigorous based on variables such as heart rate and MET value.
• Formulated multiple hypotheses and tested them using various statistical tests such as the Z-test and the Student's t-test.
• Implemented the K means clustering algorithm to achieve high accuracy in activity recognition and evaluated the performance of the model using the Silhouette coefficient.

Traffic Flow Modelling via Totally Asymmetric Simple Exclusion Process (TASEP)

• Conducted an in-depth research study on traffic flow modelling using TASEP to gain insights into traffic dynamics and optimise transportation systems.
• Developed a comprehensive literature review to establish the theoretical foundation of traffic flow modelling, highlighting key models, methodologies and research gaps.
• Collected and analysed real-world traffic data from the UK Government's website and utilised variables such as traffic speed and density to calibrate and validate the TASEP model against empirical traffic behaviour.
• Analysed large scale traffic scenarios efficiently using simulations and compared them with the traffic data to validate the TASEP model's capability to predict traffic flow, congestion and queue formation accurately.
• Highlighted strong analytical and problem-solving skills by addressing challenges related to traffic flow modelling, such as model calibration, capturing lane-changing behaviour etc. and further demonstrated mathematical proficiency by utilising concepts such as stochastic processes, goodness-of-fit tests, Markov Chains, PDE and various advanced mathematical proofs.

Monte Carlo Simulations

• Developed and implemented Monte Carlo simulations and Markov Chain analysis to study the Ising model, examining properties such as energy per spin, magnetisation per spin, and comparing the theoretical results with the simulated outcomes.
• Developed an algorithm in R to generate random numbers from Poisson distribution, represented the values in the form of a histogram and superimposed a theoretical curve to visually compare the empirical distribution with the theoretical distribution.
• Computed the invariant distribution of a Markov Chain, demonstrating a deep understanding of Markov Chain theory, and developed a Monte Carlo program to simulate the Markov Chain and validate the theoretical invariant distribution through empirical analysis.
• Implemented the Metropolis algorithm to generate Monte Carlo simulations of an Ising model, thereby validating the theoretical results obtained through rigorous mathematical calculations.