We aim to capture the fine-grained opinions on social trading platform, and attempt to summarize the textual data into indicators. With the thorough experiments, we find that the opinions of individual investors are comparable and complementary to that of analysts. The success of the extracted opinions on market movement prediction evidence that seizing the fine-grained information is promising.

We demonstrate the first real-time fine-grained opinion indicator (OI) generation system for traders with five novel OIs, which are constructed by using the textual data crawled from Twitter. The empirical studies show that the proposed OIs are leading indicators for stock price movement prediction, and contain the additional information for both short-term and long-term trading.

The demonstration system won the second prize in the FinTech competition. That shows the potential value of our system and also indicates that the proposed system is promising. We organized a shared task, and attracted 12 teams from both industry and academia. That shows the proposed task is important and worth paying attention. Several researches can be developed basing on our pilot work.