You will get Quantification for estimating class distribution or prevalence
Project details
Quantification is an emerging supervised learning task in the recent decade that aims to predict the class proportion in the test sample. It has some similarities with classification but their final objectives are different. It can be employed in several applications where the demand is to understand the behavior of groups. For instance, in sentiment analysis, the interest is in the estimate of positive or negative class ratios (quantification task) instead of classifying individual predictions (classification task).
This project can be applied in health sciences to estimate the class distribution of harmful mosquito species e.g. Aedes Agypte that spread Dengue.
This project can be applied in health sciences to estimate the class distribution of harmful mosquito species e.g. Aedes Agypte that spread Dengue.
Machine Learning Tools
Keras, NumPy, pandas, Python, Python Scikit-Learn, scikit-learn, TensorFlowWhat's included $30
These options are included with the project scope.
$30
- Delivery Time 3 days
- Number of Revisions 1
- Number of Model Variations 5
- Number of Scenarios 4
- Number of Graphs/Charts 10
- Model Validation/Testing
- Model Documentation
Optional add-ons
You can add these on the next page.
Data Source Connectivity
+$20About Waqar
Machine Learning Engineer | Python, GNNs, Data Science & XAI
Sao Paulo, Brazil - 3:17 am local time
I hold a Ph.D. in Computer Science (Data Science) from the University of São Paulo and currently work as a Postdoctoral Researcher specializing in machine learning, graph neural networks, spatio-temporal modeling, explainable AI, and visual analytics.
I can help you with:
• Python data analysis and preprocessing
• Machine-learning model development and evaluation
• Graph Neural Networks using PyTorch and PyTorch Geometric
• Graph construction and representation learning
• Time-series and spatio-temporal prediction
• Explainable AI and model interpretation
• Dimensionality reduction and interactive visualization
• Research-paper implementation and experiment reproduction
• Large-scale ETL and data-processing pipelines
• Technical reports, documentation, and scientific writing
My previous work includes developing spatio-temporal GNN models for urban prediction, integrating heterogeneous datasets into graph representations, building explainable-AI and visual-analytics tools, and creating distributed ETL pipelines processing more than 100 million records per day.
I have published research at international venues including IEEE ICDM, IJCAI, IEEE DSAA, BRACIS, and SIBGRAPI, and received the IEEE DSAA 2020 Best Paper Award.
Technologies: Python, PyTorch, PyTorch Geometric, TensorFlow, Scikit-learn, Pandas, NumPy, SQL, Apache Spark, HPCC Systems, MLflow, Plotly, Matplotlib, Tableau, and Git.
You can expect clear communication, carefully documented code, reproducible results, and an honest assessment of what is achievable with your data.
Steps for completing your project
After purchasing the project, send requirements so Waqar can start the project.
Delivery time starts when Waqar receives requirements from you.
Waqar works on your project following the steps below.
Revisions may occur after the delivery date.
Data preprocessing and cleaning
Removal of Null values fill empty values according to requirements Standardize data Normalize data if required or on demand
Exploratory data analysis
To insight data and take meaningful full information EDA will be employed. In detail EDA analysis will be done