WEB-BASED ARTICLE SUMMARIZATION WITH MACHINE LEARNING TECHNIQUES
The motivation behind this project is the increasing amount of information available on the internet, which makes it difficult for people to sift through and find the relevant information they need. Text summarization can help to address this problem by condensing lengthy texts into shorter su...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Year Project Report |
| Language: | English English |
| Published: |
Universiti Malaysia Sarawak, (UNIMAS)
2023
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/44163/1/Lim%20Wu%20Tong%20%2824%20pgs%29.pdf http://ir.unimas.my/id/eprint/44163/2/Lim%20Wu%20Tong%20ft.pdf http://ir.unimas.my/id/eprint/44163/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The motivation behind this project is the increasing amount of information available on
the internet, which makes it difficult for people to sift through and find the relevant information
they need. Text summarization can help to address this problem by condensing lengthy texts
into shorter summaries that convey the main points and ideas of the original text. However,
traditional text summarization methods often produce summaries that are too short or lack
coherence, which can make them difficult to understand. Machine learning techniques have the
potential to overcome these limitations and produce more accurate and coherent summaries.
In order to develop the web-based article summarization system, various machine learning
techniques were studied and compared. The Naive Bayes, Neural Network, and decision tree
techniques were chosen for their ability to handle both numerical and categorical data, and
their robustness to noise and missing values. These techniques were implemented using the
Python programming language and the scikit-learn library. The front-end of the system was
developed using the Django framework, along with HTML, CSS and JavaScript for styling and
interactive elements. The performance of the system was evaluated using a dataset of articles
and their corresponding summaries. The quality of the summaries was assessed using metrics
such as ROUGE and expert evaluation, while the preferredness were evaluated through user
surveys and time efficiency observed from the system. The results showed that the system was
able to produce summaries that were of good quality, preferred by users, and efficient in terms
of time. Overall, the web-based article summarization system with machine learning techniques
demonstrated the potential to be a useful tool for condensing and summarizing texts in a more
accurate and coherent manner |
|---|