Projects / Programmes
Model for Domain-Specific Trend Prediction based on Semantic Enrichment of Unstructured Patterns
| Code |
Science |
Field |
Subfield |
| 5.13.00 |
Social sciences |
Information science and librarianship |
|
| Code |
Science |
Field |
| P175 |
Natural sciences and mathematics |
Informatics, systems theory |
| Code |
Science |
Field |
| 5.08 |
Social Sciences |
Media and communications |
trend prediction, patterns, semantic enrichment, integration
Researchers (13)
| no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
| 1. |
16154 |
PhD Marko Bajec |
Computer science and informatics |
Researcher |
2013 - 2016 |
490 |
| 2. |
34116 |
PhD Robert Dukarić |
Computer science and informatics |
Researcher |
2014 - 2015 |
19 |
| 3. |
30128 |
PhD Aleš Frece |
Computer science and informatics |
Researcher |
2013 - 2016 |
29 |
| 4. |
18337 |
PhD Branko Matjaž Jurič |
Computer science and informatics |
Head |
2013 - 2016 |
709 |
| 5. |
35581 |
PhD Andrej Kocbek |
Computer science and informatics |
Researcher |
2014 - 2015 |
7 |
| 6. |
30942 |
PhD Marcel Križevnik |
Computer science and informatics |
Junior researcher |
2013 |
30 |
| 7. |
12570 |
PhD Dunja Mladenić |
Computer science and informatics |
Researcher |
2013 - 2016 |
662 |
| 8. |
15087 |
PhD Mihael Mohorčič |
Telecommunications |
Researcher |
2013 - 2016 |
476 |
| 9. |
29002 |
Blaž Novak |
Computer science and informatics |
Researcher |
2013 - 2016 |
17 |
| 10. |
32697 |
Martin Potočnik |
Computer science and informatics |
Junior researcher |
2013 - 2015 |
12 |
| 11. |
32441 |
PhD Aleksandra Rashkovska Koceva |
Computer science and informatics |
Researcher |
2013 - 2016 |
82 |
| 12. |
30949 |
PhD Sebastijan Šprager |
Computer science and informatics |
Researcher |
2014 - 2016 |
69 |
| 13. |
06875 |
PhD Roman Trobec |
Computer science and informatics |
Researcher |
2013 - 2016 |
469 |
Organisations (2)
Abstract
Recent technological advances made a great step towards the possibility of forecasting trends based on the semantic enrichment of unstructured patterns. With the expansion of the Internet new sources of mostly unstructured data constantly arise. However, the field of methodological analysis of these data with the goal of pattern recognition and trend forecasting is still underdeveloped. The results of trend forecasting based on simple searches in search engines are surprising and show that the potential is huge. With the help of Google Trends, researchers showed the correlation between searches by company names and the turnover on the stock exchange; and between the content of the searches and the movements of countries GDP. Moreover, they also developed approaches for determining the expected inflation rate and unemployment, for evaluating the current sales volume at the state level, etc.
On the other hand, a significant progress in the field of analysis of large amounts of unstructured data has contributed to the successful extraction of formal knowledge from this data. Due to abundance of such data and the absence of adequate methodological support, pattern recognition and trend forecasting is still too demanding of both time and financial terms. At the same time advances in cloud computing, processing large amounts of data (Big Data) and large number of transactions (XTP - Extreme Transaction Processing) allows the development of such solutions without building costly data centres.
We believe that it is possible to develop an automated model that will not only recognize patterns, but will also be able to use them to forecast trends within a particular domain by leveraging methods of data acquisition, analysis and data sampling from heterogeneous data sources. This will include web site and portal search results, email messages and posts in social networks. The model will in an innovative way use the methods of statistical analysis and business intelligence, especially OLAP (OnLine Analytical Processing) and data mining, supported by cloud computing, Big Data and XTP. Our approach exploits the existing models for obtaining formal knowledge, introduces an innovative consensus-based decision model for pattern recognition using methods of artificial intelligence and an innovative mathematical model for trend forecasting. The proposed common solution can be adapted to particular domains, which can provide greater relevancy and accuracy of forecasts in shorter time with fewer resources as the proposed approach uses domain-specific input data.
The main objectives of the project are to develop the following deliverables, which also serve as the basis for assessing success of the project. We will develop: (1) common contextual model for pattern recognition, (2) common contextual model for trend forecasting, (3) system prototype to demonstrate the feasibility of the proposed approach in domain-specific environments: in state defence activities (MORS - Ministry of Defence of the Republic of Slovenia), in telecommunications (Telekom Slovenia d.d.) and in case of the bigger project also in the electricity consumption domain (Informatika d.d.).
The proposed project addresses fundamental challenges of business intelligence on data generated both within organization and in wider-society. Analysis of existing work has shown that the proposed project presents a unique and innovative approach on the world-level and builds on lessons from EU/FP7 projects in which both applicants and collaborating partners from EU have participated. For the purpose of development, verification and validation of the proposed model we will leverage the computer cloud at the UL FRI Cloud Computing Centre and CLASS competence centre. The project represents a continuation of the project L5-2245 and several EU/FP7 projects in which applicants participated. The project is being applied under the competence centre CLASS.
Significance for science
In this project, we have developed a modular, multi-phase, consensus-based decision model for pattern detection and trend prediction from (un)structured data, which represents an innovative contribution in many research areas. On one hand, these include activity in the area of acquisition and interpretation of unstructured data, text- and data-mining, data enrichment and acquisition of formal knowledge and on the other hand, activities in the fields of pattern recognition, artificial intelligence, time-series processing and machine learning. By using the proposed model and its associated components, researchers in the field will be able to study options for improving their results in relation to the individual models according to the related research activities. Results of the project also open many possibilities for further research in these fields. Research results have also showed great importance in the rapidly developing field of business intelligence on the basis of (un)structured data. With the support of “what-if” scenarios trend prediction can be directly included into decision-making process and operations of predictive analysis. We expect for our model to further consolidate pattern detection and trend prediction as a perspective research field and to accelerate its growth. Verification of the model in specific domains (electro-distribution) is expected to speed up the flow of the knowledge from researched area to the businesses. In this way, it may facilitate additional investments of private funds in this particular area. We expect for the proposed approaches to be one of the factors which will play an important role in the process of digital transformation, which currently represents the most important paradigm in the terms of the integration of ICT, business world and society. The proposed modular multi-phase model also allows integration into existing IT systems via service interfaces, enabling the collection and processing of unstructured data from IoT devices. Integration approach also makes use of the innovative methods for analyzing unstructured text samples. Innovativeness of the new proposed integration approach is also represented by a specific domain ontology and innovative use of the artificial intelligence, machine learning and processing time-series for iteratively improving them through their transformation for improvement of the quality of gained knowledge. Project results also represent an upgrade of the results of the EU/FP7 project RENDER, X-Like and MetaNet, and content wise represents the continuation of the project L5-2245 and competence center CLASS. The project also discussed and upgraded certain aspects of the Initiative NESSI (Networked European Software and Services Initiative): engineering services and systems, adaptive interaction, openness, integration of the company with the technology and direct-trust, security and dependency. The concept of the project and its results meet the objectives and policies of ICT within H2020, which allows for the continuation of research activities and application proposals for new EU projects.
Significance for the country
Results and solutions, developed in this project, will play an important role as a part of digital transformation, which is currently the most important paradigm in the context of comprehensive integration of society and businesses with ICT. Its realisation presents a big challenge, including effective convergence of process, business and technological aspects. In the context of digital transformation, project will have special and lasting meaning, especially for local small- and medium-sized businesses (SMEs), since it will enable them to predict trends and perform data analysis, that could not be performed before, due to the high cost. Therefore, businesses will be able to identify new niche markets or significantly improve their competiveness in the sector, all of which could accelerate their development. Project will also have special and lasting meaning for organisations, which will be able to use the results of this project to improve their communication with the general public. Project group will be able to use concrete solutions and results, developed during this project, in the process of realising digital transformation in Slovenia. A part in this is played by the project DIGITRANS, which aims to accelerate trans-regional realisation of digital transformation of SMEs in the Danube region, is expected to help in developing new and improving existing SMEs in Slovenia, and will enable their efficient integration in the region and better global reach. Proposed solutions present key tactical and strategic tools for organisations, used for predicting consumer satisfaction with provided activities, products or services and defining new strategies and future activities in the organisation. Results of the project can be seen as a key tool for improving competitive advantage of businesses and therefore an important tool for improving their market position, which is directly reflected in financial efficiently and presents an important tool for improving business efficiency, especially in the period of business recovery. Results of the project can also be used for more efficient communication between organisations and general public in order to gain better public support for their activities. That can be achieved in term of performance indicators and metrics, quality provisioning and efficient trend prediction, and risk assessment. Realised solutions will also have a big impact in the field of API ecosystems, which present an important novelty in context of enterprise systems and play one of the key roles in digital transformation. With the help of advanced analytical tools and visualisation, API ecosystems enable efficient insight and understanding of the process of developing enterprise strategies and ecosystem dynamics. Solutions and knowledge, obtained during this project, will therefore also play an important role in managing APIs, especially in the processes of API control and monitoring. With results, applicable in real-world application domains, realised project addresses goal “Increasing society’s prosperity with higher employment rate, higher added value, higher productivity, higher happiness index” of the emerging “Strategy for Slovenia’s development in the period 2013-2020”. For example, results in the domain of electrical energy consumption address the goal “Providing lower emissions and more energy from renewable sources” from the same strategy. Realised project also addresses the priority field “Sustainable growth – promoting competitive and green business with more economical resource consumption” of strategy “H2020 – Strategy for smart, sustainable and inclusive growth”. Concrete project results, mainly from the domain of electrical energy consumption, which was used as a pilot domain during the project course, address the goal “Twenty percent increase in energy consumption efficiency”.
Most important scientific results
Annual report
2013,
2014,
2015,
final report
Most important socioeconomically and culturally relevant results
Annual report
2013,
2014,
2015,
final report
