Artificial Intelligence
This section contains Artificial Intelligence related projects that were done using Python. Each of those projects uses real life datasets with scalable solutions that can be used on a daily basis.
Malaria Detection
Using a Dataset from the United States Department of Health, the goal was to detect blood cells that were infected by Malaria. After trying out different AI techniques such as CNN and VGG16, the accuracy of our final model ended up being 95%. Put to scale, this can save millions of lives when used in a real world scenario by detecting infected cells more quickly and more efficiently than when done by a human being.
Movement Recognition
Artificial Intelligence was used in this project to recognize different human movements such as push ups, lunges and jumping jacks. To do so, different layers of a CNN model were optimized to minimize errors and maximize efficiency. This model managed to identify a never seen before video and properly classify the movement that was being filmed.
NLP to Summarize Text
With the use of Natural Language Processing and a similarity matrix, online articles can be summarized into one sentence. Sentences were changed into vectors, with the one carrying the most information being chosen as the summary phrase. This technique can be used to shorten big paragraphs, or even small books.
