Skip to main content

[ui!] DATALABs - city data analytics and machine learning

[ui!] DATALABs is our comprehensive big data analysis solution that enables cities to manage their data streams with a flexible and powerful tool set, leveraging deep integration with the [ui!] UrbanPulse data platform. Our Data and Analytics team can provide your city with [ui!] DATALABs to provide powerful insights from your data.

The [ui!] data analytics and machine learning team consists of leading data professionals, many of whom have worked in national and international implementation and consulting projects with clients from municipalities, cities, or industry. Under the overall leadership of Prof. Dr. Lutz Heuser, Managing Director and Founder of [ui!], experts Dr. Manuel Görtz (Director Analytics Products), Dr. Stefan Radomski (Head of Data Analytics) and Dr. Nick Frangiadakis (Principal Data Architect) are an interdisciplinary and experienced team of software developers, data scientist and mobility experts.

City insights

We have worked with leading cities all over the world using [ui!] DATALABs to produce deep insights including;

  • identifying the demand for parking, mobility services, traffic jams and source/destination analytics
  • air quality monitoring
  • waste management optimisation
  • water management
  • footfall and time spent at destination analysis
  • analysis of demographic data and origin of visitor flows
    • [ui!] carried out multivariate analysis classifying tourists by defined features such as distinguishing national and international visitors, visitors with and without overnight stays
  • analysis of visitors' perception of the tourist destination through social networks, opinion platforms and surveys
    • [ui!] used neuro-linguistic programming methods and sentiment analysis on external social media sources to analyse the satisfaction and the perception of the visitors
    • trend and perception analysis provided additional insight

Picture2 

[ui!] DATALABs building blocks

  • [ui!] DATALABs uses proven components and open interfaces for data exchange and connection to external data sources
    • technologically, the big data analytics capability is a combination of [ui!] UrbanPulse, our open real-time data integration and data repository, and [ui!] DATALABs Apache Hadoop service for big data analysis and fast data analysis
  • [ui!] DATALABs are based on Jupyter notebooks with access to our Apache Hadoop infrastructure as a service and enables both Big Data analysis and Fast Data analysis
    • currently we process 450 million datasets per day (around 13 billion datasets per month) and we perform about 11 million neural network evaluations for traffic light phase prediction
    • the [ui!] DATALAB enables cities to independently generate regular reports
  • [ui!] DATALAB technology integrates directly into [ui!] UrbanPulse and provides customisable visualisation of the data
    • for example, displaying the location and status of data points or analytic results on a map or to calculate and graphically display defined KPIs
    • [ui!] DATALAB visualisation provides access to a large amount of historical and current data via web browser as well as functions for customisable exploration, manipulation and visualisation
  • an important feature of [ui!] DATALAB, integrated with [ui!] UrbanPulse business intelligence solution is the control that users are provided to view the stored data
    • for example, data can be examined using drill-down and slice-and-dice analyses to draw sound conclusions. KPIs can be defined and compared with the current data
  • [ui] DATALAB can be processed, organised and categorised as .CSV export. Alternatively, exports in JSON format are also possible

Contact

If you would like to understand more about how [ui!] DATALABs could benefit your city or which cities are taking the lead in exploiting big data analysis, please contact me This email address is being protected from spambots. You need JavaScript enabled to view it.

Brief2 This email address is being protected from spambots. You need JavaScript enabled to view it.