European Commission logo
polski polski
CORDIS - Wyniki badań wspieranych przez UE
CORDIS
Zawartość zarchiwizowana w dniu 2024-06-18

Supporting SMEs innovation processes through advanced IT search systems

Final Report Summary - INSEARCH (Supporting SMEs innovation processes through advanced IT search systems)

The INSEARCH project developed an information system, based on advanced knowledge search, retrieval and analysis techniques, able to support small to medium-sized enterprises (SMEs) in the technology innovation processes through the search and usage of available knowledge present in the form of structured and unstructured data in the enterprise and in the World Wide Web.

The INSEARCH tool, through the usage of advanced filtering, recommendation, and monitoring techniques, as well as natural language processing and text mining techniques, supports SMEs to find and use information relevant for their innovation processes such as patents, technical papers, competitors and other interesting organisations web page contents, and other documents available on the web.

The implemented INSEARCH tool is a complete information retrieval (IR) and analysis system that can be used by SMEs both as a service accessible via internet and as a complete system to be installable in their premises and integrated with available knowledge management systems.

INSEARCH is composed by several modules, fully integrated in the final tool but usable also in a separated way, providing specific functionalities supporting the innovation processes, namely:

- the possibility for the users to define a domain of interest and decompose the related concepts through a tree structure, available for contextual searches, categorisation, and recommendations;
- the possibility of performing advanced searches using NLP techniques to find patterns within documents describing physical action that a 'tool' performs on an 'object', abstracting problems from the specific case under study to suggest possible innovations;
- performing searches using a part of speech (POS) approach for defining terms of interest, narrowing the search or performing searches in very general terms.

Project context and objectives:

The INSEARCH project developed a specialised information system, based on advanced knowledge search, extraction, retrieval and analysis techniques, able to support European SMEs in the technology innovation process, in particular in facing one of their main challenges: how to find and use of available knowledge (internal and external) that could be used to support the process of innovating products and processes.

INSEARCH uses advanced filtering, recommendation, and webcrawling techniques to search and retrieve relevant information, as well as natural language processing and text mining techniques to analyse the retrieved information to be translated into knowledge supporting the innovation process.

In order to satisfy the SMEs' needs and provide them with a complete and personalisable system, INSEARCH is realised both as a system usable as service, available via web, and as a system installable directly in the SME premises.

INSEARCH reached the following specific objectives:

1) A complete operational INSEARCH information retrieval (IR) and analysis system that could be used by SMEs. The measurable and verifiable final result is the presence and full availability of the system providing a set of functionalities accessible by SMEs both as a service, as well as being delivered as a system to be installable in SMEs premises.
2) The specific modules that are the core of the information retrieval system; such modules are integrated in the final system but can be used also in a separated way to provide specific functionalities within knowledge management systems; the measurable and verifiable result is the development and full availability of the following modules:

1. Webcrawler, able to perform searches in back end automatically, providing the user with suggestions on web sites and documents with a certain frequency;
2. OAT search, a component that enables the user to perform advanced researches in his knowledge base. Object action tool (OAT) is used in the theory of inventive problem solving (TRIZ) methodology with the specific aim to define the physical action that the 'tool' performs on the 'object', abstracting problems from the specific case under study to find general solutions.
3. Domain search, supporting users' contextual searches throughout the system into specific domain.
4. Semantic bookmarking, able to support the user in annotating his websites of interest, enabling the system to extract the user interest concepts and thus learn more on the user domain.
5. Personalisation, enabling the user in defining a domain in the form of 'tree of topics' which his domains of interest and creating the structured tree of related concepts, available later for contextual searches and as categories inside the advanced OAT search.

The overall INSEARCH system and the modules composing such system are parts of the complex scenario of an innovative view on how to support activities related to technology innovation in SMEs.

Project results:

The main innovative, scientific and technological results and the scientific challenges achieved during the development of INSEARCH project are related to the development of the OAT search and the possibility to personalise the domain of interest and perform contextual searches.

OAT search

The triples OAT is used in the TRIZ methodology with the aim to define the physical action that the 'tool' performs on the 'object', thus modelling technological systems by abstracting problems from the specific case under study in order to find general solutions and then apply such solutions again to the specific case under study.

Each technological system can be composed by a tool (a component realising an action), an action (the function which the component 'tool' realises) and an object (the component against which the function is realised). By focusing on the specific physical action, the user is able to abstract the problem and identify possible innovations in sectors (industrial / market and technology sectors) that not necessary belonging to the same sectors where the system under study is operating.

The big challenge achieved by INSEARCH is having developed a module able to support the application of the TRIZ OAT modelling by searching in documents for patterns that represent OAT, with the final aim to suggest to users possible innovation. This result was reached joining in an interdisciplinary way innovation management methodology like TRIZ with information retrieval and extraction technologies.

The result obtained is an information technology tool based on semantic technologies enabling the user to discover in a semi-automatic way patterns and sentences useful for suggesting possible new ways to innovate products, thereby making more automatic the application of methodologies for systematic innovation.

The users are enabled by the OAT search tool to perform advanced research in their knowledge base and to define OAT search query could by specifying terms of interest contained in the query. For each term, it is possible to define several required features which can be very specific (like e.g. word = 'coffee') to very general (like e.g. part-of-speech = 'verb'), allowing a broad flexibility for different generalisation needs of user searches.

The search for more generalised triples can be supported by the following processes:

- Lemmatisation: It allows to search for certain terms regardless their actual characteristic of the word but for terms having the same lemma. The lemmas of the actual words are retrieved in the analysis phase.
- Categories: The users can not only define their search terms by defining the string or lemma of a term but also more general by using just a category (like e.g. category ='machine'); moreover, the users can also model their own category knowledge by using the INSEARCH user interface.
- Part-of-speech (POS): The user is able to define a POS for his terms of interest, with the results to narrow the search or to define his search in very general terms.
- OAT-category: OAT categories in the INSEARCH project are 'objects' and 'actions'. Like with part-of-speech, the user can either narrow the search by setting 'object' or 'action' as additional requirement for a certain term, or he can define very general terms by setting the requirement for a term in the query just to 'object' or 'action' (e.g. 'object mills beans' to retrieve all things which mill beans).

Personalisation and contextual search

Most of the state-of-the-art techniques for semantic enterprise is included in the INSEARCH project: highly accurate lexical semantics, semantic web tools, collaborative knowledge management and personalisation.

The outcome realised is an advanced integration of analytical natural language analysis tools, robust adaptive methods as well as semantic document management over the semantic web standards.

In INSEARCH, standard models and technologies of the resource description framework 1 family (RDF 1 family) have been adapted in order to model the information associated to user management, domain modelling and user data.

These three different aspects have been physically modularised by partitioning the triples content. Each of these partitions is in turn divided into smaller segments to further account for specific data organisation requirements such as provenance and access privileges. The partitions are obtained through use of RDF named graphs, so that, whenever appropriate, the knowledge server may benefit of a single shared data space, or is able conversely to manage each partition (or set of partitions) as a separate dataset.

Users are able to access, create or refine descriptions of a domain in the form of 'tree of topics'(modelled as SKOS2 concept schemes) which support their contextual search throughout the system. These topics act as collectors for documents which expose textual content naturally associated to their definition. They are under all aspects a controlled hierarchical vocabulary of tags offered to the users.

Potential impact:

Impact for the SME participants and competitiveness

The INSEARCH project was specifically targeted to address the needs of the SMEs proposers. In order to obtain a useful and ready-to-use tool, the SMEs partners performed a validation case per each one of them, tailoring the system on their specific needs. This approach allowed SMEs to be able to use the system to innovate their products and / or to sell advanced services / solutions, thus reaching their expected main benefits.

Two of the SMEs can be considered end-users of the project results and they will utilise the developed tool to support the innovation of their own processes; the following main return on investment could be foreseen.

a) Reduction of costs related to innovation process mostly due to the support that INSEARCH can provide in the overall innovation process and to the possible elimination of costs related to outsourcing some of the activities related to product innovation to external consultants.
b) Improvement of internal capability in planning and implementing technology innovation.
c) Gather and provide relevant information on technology products to enhance internal databases.

Two SMEs are information and communication technologies (ICT) companies that will make use of the project results to improve their portfolio of solutions offered in the market, in particular by:

a) selling the tool to other organisations;
b) use the results achieved to integrate such results in the current products.

Areas of application

The consortium developed the INSEARCH information retrieval system addressing the requirements identified by the SMEs involved in the project, thus with main target group being electro-mechanical industry.

At the same time, as far as the general architecture and modules developed in INSEARCH can be adapted to the search and retrieval of information in several sectors (for market and technological purposes), the results of the project can impact on diverse sectors and areas of applications where innovation is driven by an advanced usage of technological and market information available on the web. The market analysis performed, in fact, through interviews realised with 90 organisations, outlined that the INSEARCH tool can provide interesting features to organisations of different dimensions and of several sectors, thus not limiting the application to electromechanical SMEs.

The impact of the project can be defined by the following elements:

1. The need for SMEs to have in hand (as a system usable as service and easy to be used, available via web, or as a system installable directly in the SME premises) a powerful tool able to support the search, retrieve and analysis of valuable knowledge to support technological innovation processes. It is a recognised need of SMEs to implement innovation, as EC and local governmental policies clearly states; in the overall innovation processes, the search and usage of external knowledge is a key factor as promoted by the open innovation paradigm. In such sense, the main challenge for SMEs is to search and use available external knowledge, present in the web in the form of structured and unstructured databases, to generate, design and implement technological innovation.
2. The INSEARCH impact involves different kinds of sectors; thus the opportunity to retrieve information through different kinds of search modality can take into account the diverse needs of different typologies of users. The usage of OAT search, used in TRIZ methodology has the specific aim to define the physical action that the 'tool' performs on the 'object', abstracting problems from the specific case under study to find general solutions. The OAT component also includes the possibility to search through part-of-speech (POS): the users are able to define a POS for his terms of interest, narrowing the search or performing searches in very general terms. This features was added to guarantee more flexibility during searches and allow a wide usage of the tool also to sectors where the TRIZ OAT approach is not easily applicable (e.g. an sector different from manufacturing).

Contribution to policy development

The INSEARCH project fits with the European polices related to the promotion of innovation as key driver for sustainable development and competitiveness:

- INSEARCH supports SMEs in increasing their capabilities to face market needs through innovation by an improved usage of available knowledge.
- INSEARCH also improves the transnational cooperation and knowledge transfer by supporting European SMEs in managing their approach towards innovation in collaborative environment.
- The benefits achievable through the usage of INSEARCH will impact SMEs operating in diverse sectors, and specifically will increase innovation capabilities of SMEs operating in the low-tech sectors (such as manufacturing).

Dissemination activities

The external dissemination activities performed during the project included publications and participation to events, as well as the realisation of a dedicated webinar. The following communication material was realised to increase the impact of the project:

1. A project logo was created at the beginning of the project to be used in all communication material and in the INSEARCH website. The slogan 'Researching beyond reach' symbolising the aim of the project was inserted in the project logo.
2. The project website was published at the address http://www.insearch-project.eu

The dissemination activities during the project life were focussed on participating to events, publishing papers and presenting the project expected results to industrial bodies.

The following conferences were attended, and papers presented:

- Business Meets Science, 31 January 2012, Kufstein, Austria; presentation of INSERCH project 'FOCUS on innovation for SME'.
- 'Tiroler Forschungsenquette' conference, 26 March, 2012 Innsbruck, Austria; paper presented to the scientific community with the title of: 'IT-support of innovation projects in SME (example insearch)'.
- Dies Academicus, 28 July 2012 Kufstein, Austria; presentation of INSERCH project on scientific public events.
- Sixth IEEE International Conference on Semantic Computing, 19 - 21 September 2012, Palermo, Italy; presentation with live demo with the title 'Innovation-related enterprise semantic search: The INSEARCH experience'.

Several presentations to industrial bodies were performed:

- Presentation of research result by all RTD performers and INNEN in a webinar for consortium partners and potential further industry users. The webinar were organised on 14 December 2012; the partners have chosen a webinar instead of classical workshop as the target audience of SMEs is more prone to attend a webinar (with no cost and very limited time needed), than a workshop where it is almost impossible that SMEs from different location from the one where the workshop is held could attend.
- Presentation of the project results to several industrial associations and organisation.

The project website is available at the address http://www.insearch-project.eu

The relevant contact details for each organisation that took part to the project are:

Innovation Engineering SRL
Mr Mirko Calvaresi
m.calvaresi@innovationengineering.eu

IQSER Technologies GMBH
Mr Jörg Wurzer
joerg.wurzer@iqser.net

Compano B.V.
Mr Henk Koops
hkoops@compano.com

ICA SPA
Mr Cesare Rapparini
cesare.rapparini@icaspa.it

FH-KUFSTEIN, University of Applied Sciences
Mr Karsten Boehm
Karsten.Boehm@fh-kufstein.ac.at

Institut fur Angewandte Informatik EV
Mr Gerhard Heyer
heyer@informatik.uni-leipzig.de

Universita degli Studi di Rome tor Vergata
Mr Roberto Basili
basili@info.uniroma2.it

Ciaotech SRL
Mr Paolo Salvatore
p.salvatore@ciaotech.com