A semantic-based knowledge flow system (AsIsKnown)

Published on: 11/08/2009
Document

AsIsKnown (A semantic-based knowledge flow system for the European home textiles industry) was a successful project for supporting the European home textiles industry with new technologies. Nine partners from six European countries formed a well-balanced consortium of associations in the home textiles industry, home textiles producers, retailers as well as experts of data modelling, data mining and ontologies.

The project aimed at reducing overproductions and storage costs and at enabling the European market to gain advantages over Asian business competitors. Within a project runtime of three years starting at April 2006, tools for customer consulting, market segmentation, trend prognoses as well as new concepts of collaborating within the home textiles industry have been developed and tested.

Description of target users and groups

Two main target groups were addressed by AsIsKnown. On the one hand there was the European home textiles industry and on the other hand there were normal end users.

Technology solution

The visible part of the AsIsKnown system for the customer is the Virtual Interior Designer (VID) - the internet based graphical user service for virtual interior design, customer consulting and catalogue browsing, developed by Crystal Design (CD, Italy). The Trend Analyser (TA) as a tool suite which supports marketing staff and integrates data across the industry's organisations reads all the data regarding orders and customer behaviour that has been collected by the VID via the DW. This is done in order to detect and evaluate producer independent trends in the European home textiles sector. As a result of this analysis, the most common combinations of product-classes, the combinations which have been viewed for the longest time, and the combinations which have actually been ordered, can be identified. For more, digitalised home textile magazines can be continually analysed in order to discover which relevant words occur most frequently and the combinations in which they appear. The TA is designed by Fraunhofer Institute for Applied Information Technology (FIT, Germany).

In order to identify interior trends and to generate new style worlds, the Smart Profiler (SP) collects all the data coming from the VID, the TA and the producers (product data). By means of the data that has been transferred, the SP is able to sense which patterns suit each other, which agegroup prefers which design, and which interior decoration is favoured by a particular group of people. The SP transfers all information to the VID enabling it to deliver targeted and up-todate design proposals to the customer. In this way it acts as a digital assistant, supporting customers as they shop home textiles. The SP samples configurations based on the producers' data on patterns, colour, quality data, prices etc. in order to develop style worlds. The SP's main innovation is the dynamic profiling using trend rules and ontologies. It is provided by BOC Information Technologies Consulting GmbH (BOC, Austria).

Further to this, AsIsKnown requires a system which can harmonise the product data of the different producers and present them to the VID in a unified way. In order to achieve this, the Converter Administration System (CAS) is used. The system is developed by RWTH Aachen University (RWTH, Germany). It delivers the different producers' data in a standardised form (XML) and therefore enables the system to offer a large combination range of different home textiles from various producers.

The ontologies, developed and managed by Institute for Parallel Processing, Bulgarian Academy of Sciences (IPP-BAS, Bulgaria), work as common language of the System. The ontologies can be seen as modelled knowledge structure for the domain of home textiles that connects the various synonyms of a notion to the same relevant concept for example. The AsIsKnown system deals with a substantial amount of information and generates new knowledge (about trends for example). In order to manage all the data that has been collected and generated, the DW is implemented. It processes, saves and administrates the entirety of the system's information.

Technology choice: Open source software

Main results, benefits and impacts

Benefits for the Producer

  • Enhanced appearance on the market
  • Professional customer service and trader service
  • Consolidated administration of data-, media and documents of each kind and use independent of time, space and place of work
  • High possibility of saving by automatic actualisation of data
  • Minimal assignment of personal in administration, economy of scale
  • Savings through less enquiries from wholesale
  • Faster coordination between persons in responsibility for products, marketing and distribution

Benefits for Retailers

  • Simplification of customer consulting and faster reply on customer requests
  • Raising quality of consultation for sale, reduced training costs
  • Easier capturing of individual customer commissions
  • Avoidance of queries and wrong orders
  • Error-free orders prevent reclamations, reduce costs and shorten delivery times

Benefits for the End-Customer

  • Collections can be presented in a more sales promoting way
  • Interested people can inform themselves, create individual wishes and experience them emotionally
  • Higher confidence in making decisions also for laymen, other people can be easily included in decision making
  • Easy and intuitive handling
  • Minimal system requirements

Lessons learnt

During the entire project period the AsIsKnown consortium successfully progressed with research activities in semantic ontologies, trend-analysing and integration of different software-modules.

Scope: International