Our Data Management Technology team works on relational database management systems, data warehouses and business intelligence systems.
Frequent sample mining
Data mining is all about finding potentially useful, previously unknown, hidden relationships in large data sets. Frequent pattern mining is a sub-task of this complex task, which involves finding different frequent patterns (sets of elements, sequences, and possibly graphs) in a large data set in an efficient way.
Clustering
Clustering is the process of grouping a set of objects into clusters such that objects in the same cluster are similar along a given similarity measure, while objects in different clusters are substantially different.
Clustering is essentially an unsupervised learning process whereby the system can automatically recognize related objects and objects that are definitely not related.
Web data mining
Within data mining, a distinct area is web data mining, where data is found on the web in the form of either web pages, or the documents and images they contain, or large data sets generated by the construction and use of web pages (web logs).
One of the main lines of web data mining is the examination of log files stored by web servers and/or clients in order to extract hidden information. This can be used, for example, to identify and group users, to restructure the website based on the knowledge gained from the use of certain web pages, etc.
Multidimensional data visualisation
Visualisation is no more than the transformation and display of data stored in numerical form into graphical form. Understanding the relationship between two variables is easier for humans if we can plot it on a graph. The simple line graph has been used for hundreds of years to represent data. The relationship between the three elements can be partially understood by creating a quasi three-dimensional view. However, understanding relationships with more than three variables using traditional techniques has limitations. Information visualisation is a new trend in data mining, aiming to develop usable multidimensional visualisation techniques.
