We build up chatbots for commercial companies to serve their needs, such as customer services. Within the whole chatbot building platform, there is one core component which is the short text similarity calculation component. We would like to improve our calculation capability for matching similar questions, as well as recommend related questions for the customers while they are chatting with the customer service agents.

Question Answering (QA) system automatically answer questions raised by users in natural languages, and it is a crucial component of a human-machine conversation system. A typical QA system collects human written question-answer groups and structures them in a database system. However, in order to answer questions that are semantically similar to the questions stored in the database but are worded differently, the QA system needs to be able to calculate the semantic similarity between different questions.

The objective of the project is to design a system that is able to generate context-wise reasonable and meaningful responses to open-domain conversation queries. In open-domain conversation generation, the retrieval-based methods and neural network generative models are two main approaches; there are also some recent research about improving the context consistency of conversation generation.