Project: Classify Query Intention

Highlights:

• The goal of this project is to classify user's query into four categories of intention.
• This model was built with CNN, RNN (LSTM and GRU) and Word Embeddings on Tensorflow.

Task:

• To identify the searching intention according to the query sentence typed by users in academic search engine.
• The intention has four main categories: topic, paper, expert and venue.

Data:

• Input: Query

• Output: Category

• Examples:

  Query	Category
  what's the most popular research area in data mining?	topic
  the papers about RL published in AAAI	paper
  researchers who're experted at deep learning	expert
  what's the top conference about databases?	venue

Prepare Data:

• We generate training data using mannually defined templates using the following commands.
• cd templates
• python3 scholar_simulator.py ../data/train.csv

Train:

• Command: python train.py ./data/train.csv

Predict:

• Command: python predict.py

Reference:

• Implement a cnn for text classification in tensorflow