With the deecoob insight platform, we focus our professional and technological expertise on development and operation of software for intelligent analyses and automated processing of very large data, information and text volumes for our customers. For this purpose, our data consultants are familiar with state of the art text mining algorithms, methods and tools. If necessary, we also implement our own algorithms and tools. Intensive cooperations with universities, scientific institutions and industry partners help us to stay up to date in this very challenging area. For our customers, we plan, implement and operate deecoob insight – a highly complex, scalable software system.
The main focus of our work is the analysis and extraction of data and information from structured or unstructured texts. Therefore we are specialized in:
→ Text & content collection
→ Text mining and text analysis
→ Content analysis
→ Data quality
→ Data automation
→ Visual analytics
→ Software development & application operation
The benefit of our platform deecoob insight and our expertise in data and text mining are very diverse. The following examples give a brief insight:
→ Web crawling, enterprise search, topic search
→ Uncovering the use of copyright protected works
→ Uncovering fake news, rabble-rousing, mobbing
→ Forensics, analysis of social platforms, portaIs, blogs
→ User monitoring and usage monitoring on the internet
→ Image, social, media, network and campaign monitoring
→ Geodata visualization / geotargeting
Our customers´ requirements and topics of our work are ambitious and challenging. Our claim is to leave well-known paths in search of new, innovative solutions. If existing processes, frameworks, algorithms or plug-ins are not suitable for optimally implementing requirements, we implement our own tools and workflows. Cooperations with universities and technical colleges are very important for us. Together with professors, lecturers and students, we develop ideas and implement them in long-term research projects.
So far, we have been working closely with the Dresden Technical University, the University of Technology and Economy Dresden (HTW) and the University of Applied Science Mittweida. Further research projects and cooperations are in preparation. In addition, we always look after students, accompanying them during their education, as well as the final diploma (Bachelor’s or Master’s Degree). As a result, we are constantly recruiting very well trained new employees from the university for our team.
Together with our partners, we are currently working on the research project “VANDA Visual and Analytics Interfaces for Big Data Environments”. The goal is to develop big data driven interfaces, which allow efficient access to strongly growing and differently structured data sets. This does not only involve processing of large amounts of data (volume) but also entails challenges in quality and speed of data processing, which can be summarized in the 4 V’s: volume, e.g. data volume in the terabyte to petabyte range; velocity, e.g. speed, in which data is generated, processed and analyzed; variety, e.g. diversity, which comprises different data types, sources and structures, and veracity, which describes the correctness and authenticity of data. This makes it increasingly difficult to process information in real-time (back-end) or make it accessible to the user in a suitable form (front-end).
The project duration is from August 2016 to January 2019. The research project is funded by the European Regional Development Fund.
Together with our partner we are currently working on the research project “RECAM”. This is a cooperation project between deecoob Technology GmbH and the Forensic Sience Investigation Lab at the University of Applied Sciences Mittweida. The abbreviation “RECAM” stands for retrospektive event monitoring for computer forensic enlightenment of copyright misuse on the basis of publicly available digital media.
The goal is to develop a technology for automated, computer-forensic investigation of use and misuse of copyrights at music events. The application uses web crawlers and full-text indexers and provides interfaces with Facebook, Instagram, Twitter, epaper, and Web pages. The purpose is to analyze, process, and qualify textual data from publicly available sources.
One basis for this is the development of a text mining method based on Elasticsearch search engine technology. An automated multi-vector space-based retrieval system should be able to recognize similar documents that contain content relevant to copyright. Variable cascading information filters increase retrival quality. This is where the LSI (Latent Semantic Indexing) method comes into play. A novel dynamically growing document clustering method is used for the first time in the field of text retrieval. It ensures that no processing and storage of personal data (to be protected by data law) takes place, since no personal metadata of a document is transferred to the system.
The project runs from September 2017 to August 2019. The research project is funded by the Central Innovation Program for SMEs of the Federal Ministry of Economics and Energy.