Fake News Detection System

def prune_transformer(model, amount=0.3): for name, module in model.named_modules(): if isinstance(module, torch.nn.Linear): prune.l1_unstructured(module, name='weight', amount=amount) prune.remove(module, 'weight') return model # Efficient Multilingual NMT class LightTranslator(nn.Module): def __init__(self, vocab_size, d_model=512): super().__init__() self.encoder = TransformerEncoder(d_model) self.decoder = TransformerDecoder(d_model) def forward(self, src, tgt): memory = self.encoder(src) return self.decoder(tgt, memory)

Project Overview

Developed a system to detect fake news articles by combining rich semantic text representations with deep neural classification models, at CERIST. The goal was to move beyond simple keyword or source-based heuristics toward representations that capture meaning and framing.

Key Features

Semantic Embedding: Word and sentence embeddings used to build dense representations of article content rather than relying on surface lexical features.
Classification: Deep neural network architecture trained to distinguish credible from fabricated content based on these representations.
Evaluation: Testing on curated datasets of labeled news articles, with attention to generalization across topics (a common failure mode for fake news classifiers trained on a narrow topic set).