New Detection and Sentiment analysis
With the growing era of social media, it is difficult to identify the validity whether it is any news or face/video of any celebrity, politician etc. Also, manipulated news are being generated enormously which are harder to detect by traditional means of software or methods. Therefore, NLP which is
2025-06-28 16:28:39 - Adil Khan
New Detection and Sentiment analysis
Project Area of Specialization Artificial IntelligenceProject SummaryWith the growing era of social media, it is difficult to identify the validity whether it is any news or face/video of any celebrity, politician etc. Also, manipulated news are being generated enormously which are harder to detect by traditional means of software or methods. Therefore, NLP which is a subset of Machine Learning can be employed to identify the realness or fakeness efficiently
We are creating a website which can detect two things.
- News validity,
- Sentiment analysis of news.
social media there are many propagandas are spread by using this website they know what is valid or what is Invalid.
Sometime we do not get the sentiment of speaker through AI we can detect the sentiment of speaker.
Project Objectives The project claims the following Objectives and Contributions:- It can detect news authenticity through its body text with 97% accuracy.
- It can search through 5 Pakistani news channel websites for checking its authenticity
- It can analysis news content where its positive negative or neutral through it headline
The project claims the following Project Scope:
- For detection we first use TF-ID vectorization on text then use online passive aggressive classifier for detection.
- For Searching we first scrap data from website then use TF-ID for text vectorization and make a search engine.
- For Sentiment analysis we use rule-based technique.
The main motivation of this website is to identify reality of any news or its sentiments. In this era where everything is digitalized it is hard to tell the difference between realness or fakeness and sometime, we get distracted by Invalid news rooming around social media. The main goal is to analyze news and extract all its aspects so there will be no confusion occurs when you heard any news.
Technical Details of Final DeliverableLINGUISTIC APPROACHES:
The linguistic approach deals with different main steps:
? Training set creation:
The training set is the material through which the computer learns how to process information. Machine learning uses algorithms. The idea is that because the machine learning program is so complex and so sophisticated, it uses iterative training on each of those images to eventually be able to recognize features, shapes and even subjects such as people or animals. The training data is absolutely essential to the process – it can be thought of as the “food” the system uses to operate.
? Feature engineering:
Feature engineering is the process of using domain knowledge of the data to create features that make machine learning algorithms work. If feature engineering is done correctly, it increases the predictive power of machine learning algorithms by creating features from raw data that help facilitate the machine learning process.
? Model training:
Train a model is a process of fit a machine learning algorithm able to infer a result given a set of variables called predictors.
The core activity in this approach is on feature engineering step, based on an exploratory analysis on the identification of linguistic differences in fake and legitimate news content.
The strategy is to build a fake news detection model, extracting a set of linguistic features and considering just features not related directly with words meaning; in particular, each word won’t count as a single feature, it will bring information to the model taking its syntax and stylometric value.
Unlike the most frequent approach in academic field, take always into consideration semantic words connotation in an article, using models like bag of words, term vector or other kind of vectorization.
LEXICON-BASED APPROACHE:This is a practical approach to analyzing text without training or using machine learning models. The result of this approach is a set of rules based on which the text is labeled as positive/negative/neutral. These rules are also known as lexicons. Hence, the Rule-based approach is called Lexicon based approach.
Widely used lexicon-based approaches are TextBlob, VADER, SentiWordNet.
In our project we combine the functionality of TextBlob and VADER
We first extract polarity through TextBlob then use VADER to extract second polarity of text and finally take mean of the value.
Finally apply the analysis on value.
Final Deliverable of the Project Software SystemCore Industry ITOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Peace and Justice Strong InstitutionsRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 70501 | |||
| prrinting | Miscellaneous | 2 | 1000 | 2000 |
| transport | Miscellaneous | 3 | 1000 | 3000 |
| AFOX GT730 4GB 128bit DDR3 Low Profile PCI-E Graphics Card | AF730-409 | Equipment | 1 | 20000 | 20000 |
| Ram | Equipment | 1 | 17601 | 17601 |
| m2 128 gb | Equipment | 1 | 27900 | 27900 |