The Titan project belongs to the Liito - Innovative Business Competence and Mangement Technology Programme of the Finnish Funding Agency for Technology and Innovation (Tekes). The funding period is 2008-2011.
The aim of the Titan project is to develop new information systems based on intelligent computational models to serve as decision aid for managers and stakeholders. In our project we conduct research on how a company can take advantage of their ERP-, and CRM systems and utilize information from the Internet to receive a competitive advantage. We use both quantitative and qualitative data.
We analyze the information concerning the companies’ customers and develop models and systems on how this information can be used for marketing purposes and for coping effectively with suppliers and customers. We develop models and systems for corporate managers so that they can benchmark their companies against their competitors and based on that make strategic decisions i.e. on possible mergers or going into new markets.
We analyze and develop models on how a company can become a winner on the global markets. We develop models and systems for how large amounts of data can effectively and efficiently be analyzed in organizations e.g. within taxation. As methods we use statistics and neural networks – k-means and different intelligent data mining methods like Prototype matching, and Support Vector Machine methods.
The amount of data and text has increased considerably during the last seven years and we are already talking about a future data amount of yotta bytes on the Internet. Today, many organizations struggle with vast amounts of data. Worldwide, computers have turned into massive data tombs. We can capture and store data, but it has become difficult to utilize it effectively and efficiently.
Our overall research goal is to search for, find and systematize hidden knowledge that is important for decision making in very large sets of data in organizations using data and text mining.
Systematizing knowledge using data and in particular text mining is new and demanding. We will focus on three application areas, i.e. financial benchmarking and performance analysis, corporate taxation, and customer analysis and segmentation.
We will carry out the whole system construction cycle, i.e., we analyze the problem area and find the relevant input variables, we design the system, we program it, validate it, and evaluate its performance in a realistic business or organization environment.
We will mainly use neural networks – in particular the self-organizing map as a data mining tool and the prototype matching method as a text mining tool. However, we will continuously look for other potential methods and compare the methods we use to other methods.
Hannu Vanharanta, Professor
Jussi Kantola, Post doc
Tapio Salminen, Ph.D. student
Camilla Magnusson, Ph.D. student
Pasi Porkka, Ph.D. student
Benita Gullkvist, Professor
Oana Velcu, Post doc
Ogan Yigitbasioglu, Post doc
Argyris Argyrou, Ph.D. student