An AI-assisted solution for semantic concept discovery
To prepare comprehensive and accurate reports on given topics or high-level questions, analysts have to first identify the topics, people, enterprises, and events associated with the topics or questions. This would pose a challenge for the analysts, as they are likely to miss one or more important sources of information. To help analysts overcome the challenge, the knowledge induction team at IBM Research AI (artificial intelligence), led by Alfio Gliozzo, built a solution using deep learning and structured event data. The AI-assisted solution can help analysts prepare complete reports and avoid bias based on experience.
The solution developed by the IBM team is innovative as it creates semantic embeddings out of structured event data. The core idea is similar to the popular and widely used idea of word embeddings in natural language processing. However, the IBM team has represented values instead of words in the structured data.
The powerful solution enables fast and effective semantic search across different fields in the database. For instance, a single search query will take only a few milliseconds to retrieve results based on mining hundreds of millions of records and billions of values.
Click here to read the complete article.