Research Topic prediction

Prediction Models/Research topic Prediction/README.md

@@ -0,0 +1,47 @@
+# Research-topic-Prediction
+
+### Problem Statement
+Researchers have access to large online archives of scientific articles. As a consequence, finding relevant articles has become more difficult. Tagging or topic modelling provides a way to give token of identification to research articles which facilitates recommendation and search process.
+
+Given the abstract and title for a set of research articles, predict the topics for each article included in the test set. 
+
+Note that a research article can possibly have more than 1 topic. The research article abstracts and titles are sourced from the following 6 topics: 
+
+1. Computer Science
+
+2. Physics
+
+3. Mathematics
+
+4. Statistics
+
+5. Quantitative Biology
+
+6. Quantitative Finance
+
+## Approach
+
+### Data Preprocessing:
+
+We clean the text data by removing punctuations, converting text to lowercase, and removing unnecessary characters. This helps standardize the text input for better performance during the model training.
+Feature Extraction:
+
+We extract features from the research abstracts and titles using text vectorization techniques. The two main methods used are CountVectorizer (which counts word occurrences) and TF-IDF (Term Frequency-Inverse Document Frequency), which assigns importance to words based on their frequency across documents.
+### Model Selection:
+
+We use a Linear Support Vector Machine (LinearSVC) with a multi-output classification approach. This allows the model to predict multiple topics for each article simultaneously.
+Evaluation:
+
+The model is evaluated using common classification metrics such as precision, recall, F1-score, and accuracy. These metrics give insight into how well the model performs across different research topics.
+### Prediction:
+
+After training, the model predicts the topics for unseen research articles, and the results are formatted for submission. Each prediction shows whether a particular article belongs to one or more of the six topics.
+## Project Highlights
+Multi-label Classification: Each article can be tagged with more than one topic, so the model needs to handle multiple outputs simultaneously.
+Text Processing: Effective text preprocessing and vectorization are key to extracting meaningful features from the research articles’ titles and abstracts.
+### Model Performance:
+ The model demonstrates good performance in most categories, but there is room for improvement, particularly in addressing class imbalances for topics with fewer articles.
+## Future Work
+Future improvements could include addressing the data imbalance, exploring advanced machine learning models such as deep learning techniques, and refining the feature extraction process for better prediction accuracy.
+
+This project provides a foundation for automating topic prediction in research articles, potentially enhancing search engines and recommendation systems in academic databases.