Use our data collection infrastructure to discover and acquire Web data
Insius has its own data collection infrastructure that is based on the simulation of human search behavior. This allows us to acquire the maximum coverage of relevant data. Furthermore, our systems not only index RSS feeds or treat webpages as one single article, Insius crawlers use computer-vision techniques to detect and collect multiple articles within webpages (e.g. articles within threads). Using fingerprinting technologies means that near-duplicates and content spam can be detected and filtered. The cleaner your data is, the better your analyses can be. Use InCore Collect to achieve maximum data quality and coverage. If you need to analyze the voice-of-the-customer, InCore Collect also helps you to separate editorial from user-generated content.
Use our algorithms to get insights from unstructured natural-language texts
Concept Recognition: InCore Analyze allows you to use Insius advanced algorithms to monitor which concepts, persons, locations or products are mentioned in natural-language texts. You can even define speech patterns yourself to mine texts for interesting aspects.
Concept-based Sentiment: Use Insius sentiment analysis and detect tonality down to the level of statements and words.
Auto Classification: Classify large amounts of data in seconds using Insius auto-learning classifiers. As a result, after a learning phase these classifiers can be used, for example, to detect whether a text has been written by a user, has been created by professionals (editorial content) or even what is the focal topic of the text.
Topic Detection: Detect the focal topic of your unstructured texts. Is it a customer query, product offering, or from what domain is the text originating?
Date Detection: Find dates in unstructured text no matter if they are relative (e.g. "next week") or exactly defined ("1996-12-12")