IFindIT Information Services

IFindIT® is a platform built on a native private cloud architecture, optimized for real-time applications, that combines agile, elastic storage capacity with deep search capabilities, unmatched relevance and AI capabilities. This architecture allows Creangel to offer its solutions under the SaaS model, with no client-side servers required, which eliminates all operating expenses related to technology infrastructure so you can get started quickly and scale your needs with a platform based on Big Data

IFindIT Information Services

Clearly, most organizations struggle to exploit the wealth of their content and turn it into something accessible and easy to use. At the same time, content grows rapidly, which makes the challenge greater over time.
IFindIT Information Services helps you generate knowledge from your organization’s structured data.
The IFindIT® platform is unique in its ability to access content in more than 250 file formats across the most diverse types of information sources, and it includes multiple ready-to-use connectors and “converters” to extract text and data from information systems. New capabilities are routinely added, providing greater visibility across organizations’ digital ecosystems.

IFindIT® uses Big Data techniques for agile information consumption, which allows it to ingest into the platform around 2,000,000 documents or 10,000,000 structured records per hour.
Connectors cover systems such as databases, document management systems, web portals, shared folders, proprietary applications, etc.
IFindIT® components allow you to detect and clean text, even text contained in scanned documents as images, and return as results a set of entities in many categories. For some of them, relationships between entities can even be generated. These relationships improve the understanding of context when an entity refers to several subjects.
IFindIT® has been specially developed and trained to work in Spanish. This is key because even highly formal documents contain local terms, and a significant volume of natural-language text is also subject to analysis, which makes working with a specific language even more important. This factor, combined with intelligent text extraction that uses custom dictionaries, text-cleaning algorithms and artificial intelligence to improve the accuracy of the base text, makes the precision, quality and quantity of identified data higher than in other general-purpose tools.

Once the text is available, IFindIT® cleans the data and “interprets” the different types of content fed into the platform. To do so, it combines natural language processing, dictionaries, artificial intelligence and other techniques that remove noise, produce cleaner data and enrich information. All of this serves as a foundation for accurate analytics, relevant searches and stronger machine learning.
IFindIT® uses machine learning, which is essential to turn data into relevant information and actionable knowledge. The platform learns from examples that train the built-in models to cover specific information needs, such as those listed below, or specific requirements (on demand).
All data generated by the analyses is subject to classification, clustering and other operations that turn data into search filters. Some of the data recognition and extraction capabilities are:

• Detection of person-type entities
• Detection of artists, politicians, writers and science fiction characters in SPANISH
• Association between roles, professions, job titles and types of work in SPANISH
• Detection of geographic locations, cities, countries, departments and man-made structures, including local ones
• Detection of government, commercial and service entities, non-profit companies, and medical, sports and stock exchange organizations, including local ones
• Detection of local holidays and cultural events, sporting events and popular events
• Detection of international and LOCAL products
• Detection of skills, physical addresses, phone numbers, email addresses and IP addresses
• Date detection
• Detection of numeric quantities
• Classification by similar document types