@anhassan
github

anhassan

Twitter logo Share LinkedIn logo Share Facebook logo Share

Top repositories

1

Quora-Question-Similarity-Using-BERT-Embeddings-with-Siamese-Networks-as-Feature-Extractors

Proposed a model architecture which learns to classify duplicate question pairs based on highly contextualized sentence representations. Achieved by using both individual and combined sentence embeddings based on Word2Vec, GloVe and FastText word embeddings with CNN(convolutional neural networks) and DNN(Dense neural network) Siamese networks and an array of features engineered using sentence structure, fuzzy logic , BERT sentence embeddings and BERT [CLS] token.
Jupyter Notebook
14
star
2

RF-Coverage-Optimization-using-Machine-Learning

• Data preprocessing and Data base management using MySQL. • Coverage Analysis using SVR (Support Vector Regression), ANN (Artificial Neural Network) with Keras, Tensor Flow and Outlier Detection through cluster-based approaches including K-Means clustering and DBSCAN • Network Health Estimation using Bayesian Kriging, Deep Learning and Graph Signal Processing by solving an optimization problem on graph using proximal splitting methods. • Developing a user-friendly GUI for the regularity authority for data visualization, analytics and store the results of the analysis
Python
13
star
3

Power-Transformer-Assessment-using-Generative-Adversarial-Networks-GANs-and-Classical-ML-algorithms

Synthetic augmentation of data using GANs to increase the number of data points and use of classical machine learning techniques such as SVM and ensemble methods(Xgboost) to asses the quality of power of transformer oil.
Jupyter Notebook
7
star
4

Bank-Design-Architecture-Object-Oriented-Design-

Object Oriented Design Pattern Used to Generate and Organize Data for a Bank
Java
2
star
5

ETL-CiCd-Jenkins-Pipeline-With-Docker

A python ETL based CI/CD Jenkins pipeline to automate Docker builds
Python
2
star
6

Web-Series-Time-Series-Forecasting

Performed in depth exploratory data analysis(EDA) to detect trends, seasonality in the given data and Dicky-fuller test to find stationarity. Used techniques like differencing to cater for problems like no stationarity. Implemented ARIMA and LSTMs to predict the future web traffic on distinct Wikipedia pages
Jupyter Notebook
2
star
7

Data-WareHouse-Design-using-ETL-pipeline-with-AWS-S3-Redshift

An ETL pipeline is constructed to load data from S3 bucket into a star schema based Data WareHouse hosted in AWS Redshift
Python
2
star
8

Linux-Cluster-Monitoring-Agent

Shell
1
star
9

DataInsights

Analytic Insights on Health Care Application
Jupyter Notebook
1
star