Hamed Ahmadinia

feedback

The experienced researcher teacher data analyst enthusiast

Publications

Data Science Portfolio

Data Science Timeline

“You can have data without information, but you cannot have information without data.” – Daniel Keys Morn

  • General Skills

    1- Data Visualization (80%)
    2- Data Prepration (50%)
    3- Machine Learning (40%)
    4- Deep Learning (30%)
    5- Statistical Analytics (600%)
    6- Natural Language Processing (30%)"

  • Technical Skills

    Operating system: Windows (intermidiacte), Linux (basic)
    Database/Server: My SQL (basic)
    Programming Language: Python (intermidiacte), scitkit-learn (intermidiate), D3.js (basic), R programing (basic), HTML (intermidiate), JavaScript (basic)
    Other Software/Tools: Deep Learning (basic), Machine Learning (basic)

  • Certificates

    1- Advanced Models in Smartpls (2022) credential credential
    2- Creating Models using Smartpls (2022) credential credential
    3- Machine Learning Foundations: A Case Study Approach (2021) credential
    4- NVivo Online Training (blended learning) (2021) credential
    5- D3.js Essential Training for Data Scientists (2020) credential
    6- R Essential Training: Wrangling and Visualizing Data (2020) credential
    7- Intermediate Pandas Python Library for Data Science (2020) credential
    8- Pandas Python Library for Beginners in Data Science (2020) credential
    9- MATLAB Onramp (2020) credential

Big Data Projects - Portfolio


A: Introduction to Analytics - Projects
1- Working with time series data - Arcada-Bigdata-Specialization - 2021
exploring data, calculating Median and SD for each place of measurement, categorizing data using Boolean indexing, Visualising
2- Financial time series forecasting - Arcada-Bigdata-Specialization - 2021
exploring data, creading datetime column, converting strings to datetime, data claening and resampling, seting traning and test datasets, fiting and transforming data, fiting a linner regression, measuring classification performance, Illustrate data using plotly, calculating additional features such as Relative Strength Index

B: Machine Learning for Predictive Analytics - Projects
1- Predict the critical temperature based on the features extracted - Arcada-Bigdata-Specialization - 2021
exploring data, applying different regression models from sklearnlinear including regresion, Ridge, Lasso, fiting and transforming data into the models, ploting obtained score
2- Working with credit card clients Data Set- Arcada-Bigdata-Specialization - 2021
exploring data, fit two binary classification models to predict the client's credit card default, perform a simple manual optimization for one of the default parameters (at least 5 different values) and plot the new obtained accuracy as a function of the chosen parameter
3- Working with Drug consumption (quantified) Data Set - Arcada-Bigdata-Specialization - 2021
exploring data, fit two multiclass classification models to predict two selected features out of 18, perform a simple manual optimization for one of the default parameters (at least 5 different values) for one of the previous models, fit one multiclass classification model for all the rest 16 features, run one binary classification model for 3 features out of 18, test the performance of the model
4- Working with time series data - Arcada-Bigdata-Specialization - 2021
exploring data, calculating Median and SD for each place of measurement, categorizing data using Boolean indexing, Visualising
5- Unsupervised Learning - Arcada-Bigdata-Specialization - 2021
exploring data, data normalization, dimensionality reduction, Clustering
6- Working with Sonar Mines vs Rocks dataset - Arcada-Bigdata-Specialization - 2021
exploring data, substitute the categories M and R of the last column for integers, find the best parameters on the validation set using gridsearch, plot a heatmap of the parameters, plot a ROC curve
7- Working with Cardiotocography dataSet - Arcada-Bigdata-Specialization - 2021
exploring data, find the best two models by creating a complete pipeline with different parameters and algorithms
8- Working with UCI Arrhythmia dataset - Arcada-Bigdata-Specialization - 2021
exploring data, multi-class classification, run three different types of feature selection methods (Univariate Statistics, model based, and Iterative Feature Selection)
9- Working with the dogs versus cats dataset - Arcada-Bigdata-Specialization - 2021
exploring data, extract the array of features for different number of images (N: 10, 100, 500, 1000, also 5000 and 12500) and for each value train your favorite binary classifier, using GridSearch to optimize some hyperparameters
10- Working with CoronaHack Chest X-Ray dataset - Arcada-Bigdata-Specialization - 2021
exploring data, training a machine learning model able to predic COVID-19 from chest X-Ray images
11- Working with yeast dataset from UCI - Arcada-Bigdata-Specialization - 2021
exploring data, remove, replace, impute, split data, build a outlier detection model to classify VAC from CYT, build a classifer using sample augmentation techniques to flassify VAC from CYT, try different methods and hyper paramters, report perfromance using F-1 score

C: Visual analytics - Projects
1- Visual Data Science with R - Arcada-Bigdata-Specialization - 2020
create a personalized theme, applying different visualizing methods including frequency plot, histogram, boxplot, density plot, scatter Plot, marginal scatterplot, beeswarm plot, hexagonal binning, scales and axes, flow analysis, parallel coordinate plot, dumbbell
2- Web-based data visualization with D3.js - Blockbulder - Arcada-Bigdata-Specialization - 2020
applying different visualizing methods including Bar Chart, Scatterplot, Interactive Chart, Animated Chart
3- Geoviz and maps with CARTO - Arcada-Bigdata-Specialization - 2020
Calculating cluster point, Risk Analysis, Police station and crime Analysis Earthquake magnitude Analysis

D: Machine Learning for Descriptive Problems - Projects
1- Mining information from Text Data - Arcada-Bigdata-Specialization - 2022
Randomly select 1000 abstracts from the whole dataset. Find the similar items using pairwise Jaccard similarities, MinHash and LSH (vectorized versions)
2- Text mining and NLP problems (IMDB reviews dataset) - Arcada-Bigdata-Specialization - 2022
Defining the project topic, and RNN models
3- Implement the selected model (IMDB reviews dataset) - Arcada-Bigdata-Specialization - 2022
Sentiment Analysis on IMDB Reviews using LSTM and Keras
4- Tuning the presented models and describing how and why the results get improvement (IMDB reviews dataset) - Arcada-Bigdata-Specialization - 2022
Comparing final model with one other RNN model