nethra8902 Goto Github PK

abstractive-text-summariser-with-nlp

Text Summarization recapitulate the content available in articles, research paper, news, paragraph or a piece of information. Automatic Text Summarization is made possible in Natural Language Processing (NLP) by employing two types of summarization techniques viz., 1) Extractive Summarization and 2) Abstractive Summarization. Extractive summarization generates summary by extracting the verbatim from the original passage whereas Abstractive summarization generates summary either by paraphrasing or by using new words instead of extracting the main points. A comprehensive news article and summary dataset has been chosen from Kaggle and the latter method of summarisation is employed in our coursework which calls for an abstractive modelling approach using sequence to sequence Long Short-Term Memory (LSTM) model. BLEU scoring technique has been used for evaluating the accuracy of the model owing to the extensive usage of the same for many of the models involving Natural Language Processing.

attention_keras

Keras Layer implementation of Attention

badminton-sport-analysis-computer-vision

The following parameters have been analysed for a short badminton singles match video: 1. Player1 and Player 2 are detected with bounding boxes, 2. Recording and displaying the Start time and end time of the match, 3. Displaying the number of hits made by each player , 4. Displaying the number of out of the boundary hits (AWAYs) made by each player during the course of match, 5. Shuttlecock is detected in every frame of the video with a bounding box.

to-be-deleted