Social Data Analytics Group in IBM Research - Brazil - overview
The Social Data Analytics group of IBM Research - Brazil focuses on technologies and applications based on the analysis of human-related data. We see our mission as the realization of the potential of Big human/social data in businesses and society.
Our approach is to investigate and develop technologies based both on personal data, such as social media posts, and on social data, such as the network of friends, news articles, and dialogue systems.
Conversational cognitive agents, or chatbots, are autonomous agents that use natural language to communicate with a human or other chabots to exchange information about a common topic. Our group does research and development of methods and tools to advance the state of the art on chatbots that can interact using natural language in a multi-party mixed-initiative conversation. The mental model of the chatbots is also enhanced with social knowledge extracted from social media networks and news websites. Our main industry application is a investment adviser that can exchange information about Brazilian finance in Portuguese and also can give finance estimations using natural language.
The Social Data Analytics group also applies methodologies from social sciences and related fields to try to understand the actual meaning, relevance, and significance of the patterns found in the data. At the same, we are also exploring applications which use human and social data in novel ways, how human data is best visualized, and the interaction of people with human data.