Chris Pollett > Students > Li

    Print View



    [CS 297 Proposal]

    [Del 1: Back Propagation]

    [Del 2: Word Embedding]

    [Del 3: Setup]

    [Del 4: QA System Architecture]

    [CS297Report [PDF]]

    [CS 298 Proposal]

    [Del 6: QA System]

Rewritten on Nov 14, 2017 Due to Change of Topic

CS297 Proposal

A Question Answering System on SQuAD Dataset Using an End-to-end Neural Network

Bo Li (

Advisor: Dr. Chris Pollett


Question Answering(QA) is about making a computer program that could answer questions in natural language automatically. QA techniques are widely used among search engines, personal assistant applications on smart phones, voice control systems and a lot more other applications. In recent years, more end-to-end neural network architectures are built to do question answering tasks. In contrast, traditional solutions use syntactic and semantic analyses and hand made features. End-to-end neural network approach gives more accurate result. However, traditional ways are more explainable. The Stanford Question Answering Dataset (SQuAD)[1] is used in this project. It includes questions asked by human beings on Wikipedia articles. The answer to each question is a segment of the corresponding Wikipedia article[1]. In total, SQuAD contains 100,000+ question-answer pairs on 500+ articles[1]. The goal of this project is to build a QA system on SQuAD using an end-to-end neural network architecture.


Week 1 - 4: 08/20 - 09/24Deliverable #1 : Calculation of Back Propagation
Week 5 - 11: 09/25 - 11/12Deliverable #2 : Implementation of Word Embedding
Week 12: 11/13 - 11/19Deliverable #3 : Setting Up Development Infrastructure and Downloading Data
Week 13 - 14: 11/20 - 12/03Deliverable #4 : Implementation of the neural network model in [2]
Week 15 - 16: 12/04 - 12/17 Deliverable #5 : CS297 report


The full project will be done when CS298 is completed. The following will be done by the end of CS297:

1. Calculation of Back Propagation

2. Implementation of Word Embedding

3. Setting Up Development Infrastructure and Downloading Data

4. Implementation of the neural network model in [2]

5. CS297 report.


  2. Wang, Shuohang, and Jing Jiang. "Machine comprehension using match-lstm and answer pointer." arXiv preprint arXiv:1608.07905 (2016).