my web application seeks to predict erroneous information, based on classification model. The latter is a primary interface, this is what the user sees once he opens the application. The logo and two others links, one of which will lead us to another interface with details about the creators application, are located at the top of the page. The other link will take us back to the page home. By entering the news in the input field in the middle of this page and by clicking on "Submit", the user of our web application can use this interface to verify the veracity of the information. Our application then performs the detection and presentation of results.
One of the most popular Natural Language Processing (NLP) tasks is the classification of the text (or categorization). This involves assigning a tag to a unstructured text that corresponds to a specified class. In simple terms, the clas- sification is the process of assigning a label to a text, indicating that it belongs to a specific class. To do this, we have taken various measures, which are presented in the following diagrams.
- first step to create a NLP template
- second step to create a NLP template
Our web application was able to identify the class of each new and af- file the result accordingly which shows and confirms the correct operation of our program. To test the functionality of our application, we have attri- two new, one is false and the other is true, the result is represented in the next screenshot.
-
False News Detection
-
Real News Detection
Licensed under the MIT License. See LICENSE in the project root for license information.