Samuel Kolawole's Projects
Developing machine learning models to classify high, medium and low back pain risk based on vide recording of gaits: Utilising OpenPose a computer vision technique for features extraction.
The housing dataset is used to develop a pipeline for running a simple linear regression in a pseudo-distributed manner (single node setup) using Spark ML. The following was implemented: • Created a Spark session, loaded the data, parse and display them using the apache spark ecosystem (pyspark). View "read me" to see full desciption.
Developing machine learning models to classify high and low fall risk based on vide recording of gaits: Utilising OpenPose a computer vision technique for features extraction.
Data collection, statistical analysis, interpretation and image classification of the obtained diseases images are the focus of this project, as well as the exploration of quantitative methods for analysing this data using machine learning (support vector machine, fully connected neural network, convolutional neural network, K-nearest neighbour etc
The author implemented simple rule base solution and machine learning approach for information retrieval and information extraction after which the result were analyzed.
The author implemented support vector machine for sentiments analysis and applied two feature extractions, Bag-of-Words (CountVectorizer) and TF-IDF (TfidfVectorizer), after which the results for both methods were analysed. The accuracy obtained for both methods were (BoW = 87%) and (TF-IDF = 86%).
The author implemented logistic regression and support vector machine for topic labelling and applied two feature extractions, Bag-of-Words (CountVectorizer) and TF-IDF (TfidfVectorizer), after which the results for both methods were analyzed.
Created image classifiers as well as an access model by using Python to create a process for the processing of image data. This pipeline include: • Pre-processing, feature extraction, train classifiers with extracted features and labels from train, test, and val set. • Evaluate models with extracted features from test and val set with Visualisation