Md Saiful Islam – Data Analyst | Public Profile on Dr.Job

نبذة عني

A seasoned Data Analyst with extensive experience in interpreting and analyzing data to drive growth for a range of industries. Possesses strong technical skills, including proficiency in SQL, Python, and data visualizat…

المشاريع

ChatGPT Sentiment Analysis

University of Portsmouth · https://github.com/Santo1337

المدة : 09-Jun-2023 - 03-Oct-2023

This project, titled "ChatGPT Sentiment Analysis", was conducted at the University of Portsmouth from June 8, 2023, to October 2, 2023. The primary goal of the project was to perform sentiment analysis on public opinions toward ChatGPT using advanced Natural Language Processing (NLP) techniques. My specific contributions to the project included implementing and comparing traditional Machine Learning (ML) models such as Support Vector Machines (SVM), Naïve Bayes, Decision Trees, and Logistic Regression with deep learning architectures like Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM). I also applied text preprocessing and feature extraction using Term Frequency-Inverse Document Frequency (TF-IDF) and word embeddings to improve classification accuracy. One of the major challenges I overcame was the integration of traditional ML models with deep learning architectures to achieve optimal results. The project resulted in a significant improvement in the classification accuracy of public sentiment towards ChatGPT, demonstrating the effectiveness of the methodologies used. The project details can be accessed via the following URL: https://github.com/Santo1337.

Restaurant and Product Review Sentiment Analysis

Varanda · https://github.com/Santo1337

المدة : 08-Sep-2022 - 15-Dec-2022

During my tenure at Varanda from September to December 2022, I was actively involved in a project titled "Restaurant and Product Review Sentiment Analysis", the details of which can be found at https://github.com/Santo1337. The primary goal of this project was to classify customer sentiment as positive or negative based on their textual reviews. To achieve this, I employed Natural Language Processing (NLP) techniques and compared various Machine Learning models including Support Vector Machines (SVM), Naïve Bayes, Decision Trees, and Logistic Regression. These models were evaluated using TF-IDF and n-gram features. One of the major challenges I faced was the accurate classification of sentiments due to the inherent subjectivity and complexity of human language. However, I was able to overcome this by fine-tuning the models and incorporating more nuanced language features. The project was successful, with the models demonstrating a high degree of accuracy in sentiment classification, thereby providing valuable insights into customer satisfaction and preferences. This project not only improved the company's understanding of customer sentiment but also helped in making informed decisions to enhance customer experience.

Cardiovascular Disease Prediction Using ML Algorithms

Daffodil International University · https://github.com/Santo1337

المدة : 07-Sep-2020 - 28-Jan-2021

During my tenure at Daffodil International University, I led a project titled "Cardiovascular Disease Prediction Using ML Algorithms" from September 2020 to January 2021. The primary goal of this project was to build a predictive model for cardiovascular disease detection using supervised Machine Learning classifiers. I specifically contributed to the development and implementation of various ML algorithms such as Logistic Regression, Random Forest, and XGBoost. I also conducted extensive feature engineering and hyperparameter tuning to optimize the model's performance. One of the significant challenges I overcame was dealing with imbalanced data, which I mitigated using SMOTE (Synthetic Minority Over-sampling Technique). The performance of the model was evaluated using accuracy, precision, and ROC-AUC metrics. As a result of my contributions, the model achieved an accuracy of 85%, a precision of 88%, and an ROC-AUC score of 90%. The project's details and code can be accessed via the following URL: https://github.com/Santo1337.

المهارات

لغة C# تحليلات جوجل جافا مايكروسوفت إكسل ماي إس كيو إل بوربوينت بايثون (لغة برمجة) تجميع أدوبي فوتوشوب نوسكيو إل باور بي آي لغة البرمجة آر (R) تابلو C Google Tag Manager Pandas Numpy Seaborn Plotly Git Kaggle Supervised and unsupervised learning Ensemble modelling Feature engineering Model training Hyperparameter optimizations Performance evaluation Predictive Modelling Data Cleaning NumPy Supervised learning Unsupervised learning Hyperparameter optimization Predictive modelling Data cleaning Machine learning NLP Natural Language Processing SVM Naïve Bayes Decision Tree Logistic Regression Random Forest XGBoost CNN RNN LSTM TF-IDF Word embeddings OpenCV Cross-validation Feature selection Bagging Deep learning Data preprocessing Text preprocessing Image preprocessing