Skill set
Industry fields of expertise
Shakib in a few words
Hello,
I am a data scientist with over 5 years of experience delivering more than 30 successful projects. I've crafted predictive models and advanced analytics solutions for both cutting-edge startups and well-established companies. I specialize in building robust data pipelines, machine learning algorithms, and deep learning models that turn complex datasets into actionable insights, while also advising on data visualization and user-centric reporting.
Here are a few of my top projects:
1. NER for Swedish Historical text
- Developed a Named Entity Recognition model tailored for the Swedish language, achieving over 90% precision in identifying key entities.
2. Wind Turbine Detection From Satellite Images
- Designed a model which got 92% accuracy to detect wind turbines from satellite images.
3. Crime Rate Forecast for Indiana state
- Built a forecasting model using time series analysis and machine learning to predict crime trends accurately.
- Enhanced prediction accuracy by 15%, aiding law enforcement resource allocation.
4. Credit Scoring Model for Bank
- Improved credit decision-making processes for banks with a 25% increase in predictive reliability.
5. Breast Cancer Detection
- Developed a deep learning model for early-stage breast cancer detection, achieving a detection accuracy of approximately 93%.
6. Apartment Sales Prediction
- Constructed a regression model to predict apartment sale prices using comprehensive market and geographic data.
- Enhanced price prediction accuracy and reduced error margins by 20%, benefiting both buyers and sellers.
These projects highlight my commitment to leveraging data science to solve real-world challenges. I'm passionate about transforming raw data into strategic insights that drive impactful decisions—let's explore how we can turn your data into your greatest asset!
Kind Regards,
Shakib Kaiser
Experience
- Freelancer.comFreelancer (Data Scientist)SOFTWARE PUBLISHINGJanuary 2025 - Today (4 months)The projects I completed:1. Crime rate detection and prediction in Indiana State using Time series and Machine learning.2. Enhance satellite images resolution up to 10x.3. Detection of Wind turbines from large satellite images through the YoloV8 model.4. Achieving 75% Nervaluate f1 score in Named Entity Recognition (NER) for Swedish historical Text identification.5. Apply Smote to increase R Squared score for a Logistic Regression classification model.6. Created KGNN model for Drug-drug Interaction prediction. Achieved 92% accuracy.7. Created a Dashboard to show and compare golf players' performances.8. Predicting Credit risk for a bank using Machine Learning.9. Create an automation app to Manage Inventory.10. Wrote a research paper in use of Deep learning in medical imagery survey.11. Python trading script integration with QuantConnect.12. Using ANN to classify breast cancer. Accuracy-99%.And many more..
- CHIRAL – CENTER FOR HEALTHResearch Intern, Health Data ScienceOctober 2023 - February 2024 (4 months)Dhaka, Bangladesh• Developed an ANN model for breast cancer tumor classification (98% accuracy) using sklearn and TensorFlow.• Collected and analyzed data from 100+ university students using Kobo Toolbox for a study on depression, anxiety, and stress.• Drafted a comprehensive research proposal for a mental health study, including methodology, analysis plan, budget, and timeline.• Ensured ethical compliance throughout the research lifecycle, including obtaining ethical clearance and maintaining data privacy.WORK EXPERIENCE
- WorldQuant UniversityApplied Data Science LabApril 2023 - April 2024 (1 year)New Orleans, LA, USA1. HOUSING IN MEXICO: Learners use a dataset of 21,000 properties to determine if real estate prices are influenced more by property size or location. They import and clean data from a CSV file, build data visualizations, and examine the relationship between two variables using correlation.2. APARTMENT SALES IN BUENOS AIRES: Learners build a linear regression model to predict apartment prices in Argentina. They create a data pipeline to impute missing values and encode categorical features, and they improve model performance by reducing overfitting.3. AIR QUALITY IN NAIROBI: Learners build an ARMA time-series model to predict particulate matter levels in Kenya. They extract data from a MongoDB database using pymongo, and improve model performance through hyperparameter tuning.4. EARTHQUAKE DAMAGE IN NEPAL: Learners build logistic regression and decision tree models to predict earthquake damage to buildings. They extract data from a SQLite database, and reveal the biases in data that can lead to discrimination.5. BANKRUPTCY IN POLAND: Learners build random forest and gradient boosting models to predict whether a company will go bankrupt. They navigate the Linux command line, address imbalanced data through resampling, and consider the impact of performance metrics precision and recall.6. CUSTOMER SEGMENTATION IN THE US: Learners build a k-means model to cluster US consumers into groups. They use principal component analysis (PCA) for data visualization, and they create an interactive dashboard with Plotly Dash.7. A/B TESTING AT WORLDQUANT UNIVERSITY: Learners conduct a chi-square test to determine if sending an email can increase program enrollment at WQU. They build custom Python classes to implement an ETL process, and they create an interactive data application following a three-tiered design pattern.
Recommendations
Education
- Master of ScienceEAST WEST UNIVERSITY2024Master of Science in Data Science and Analytics
- Bachelor of ScienceInternational University of Business Agriculture and Technology2022Bachelor of Science in Economics