Dynamic Data Mining Research Articles

A granular hardware/software system: pervasive distributed dynamic sensor data mining system for effective commerce

In recent years, wireless sensor networks, cell phones and granular hardware/software systems have become ubiquitous for various applications; fusing these technologies in the field of business opens up new applications. To fill this lacuna, we propose a novel idea where the combination of these technologies facilitates the stores to receive feedback on their products and services. System's unobtrusive sensors will not only collect shopping data from customers, but will also make effective use of this information to increase revenue, cut costs, etc.; the use of mobile agents-based data mining allows analysing the data from different dimensions and categorising it on factors such as product positioning, promotion of goods, etc. Additionally, the shoppers get on-the-scene recommendation of products, rather than off-the-scene. In this paper, a novel granular distributed pervasive mining system is proposed to get dynamic shopping information of the customers.

Dynamic data analysis and data mining for prediction of clinical stability.

This work studies the impact of using dynamic information as features in a machine learning algorithm for the prediction task of classifying critically ill patients in two classes according to the time they need to reach a stable state after coronary bypass surgery: less or more than nine hours. On the basis of five physiological variables different dynamic features were extracted. These sets of features served subsequently as inputs for a Gaussian process and the prediction results were compared with the case where only admission data was used for the classification. The dynamic features, especially the cepstral coefficients (aROC: 0.749, Brier score: 0.206), resulted in higher performances when compared to static admission data (aROC: 0.547, Brier score: 0.247). In all cases, the Gaussian process classifier outperformed logistic regression.

Combination of dynamic analysis and data mining for stability prediction after cardiac surgery

Combination of dynamic analysis and data mining for stability prediction after cardiac surgery

Intelligent cluster algorithms for changing data structures

Many real life applications are characterised by changing data structures. For example, the buying patterns of retail customers may change due to changing economical parameters (increasing oil prices motivate to buy smaller cars) or a technological break-through (replacement of analogue by digital cameras). In such dynamic environments the parameters obtained in data mining projects need to be updated to adequately describe the actual real life situation. Dynamic data mining addresses such situations. It has been applied successfully in many projects, like in traffic data analysis. In our paper, we apply the concepts of dynamic data mining to rough k-means.

Applying Dynamic Data Mining on Multi-Agent Systems

A new perspective on analysis of large-scale Multi-Agent Dynamic Systems is presented. The aim is to capture the global trends at a glance, by the use of dynamic data mining techniques, which group agents according to similar characteristics or behavior, measuring and recording how the different trends evolve through time. This methodology is presented and an example with a simulated dynamic Multi-Agent System is included.

Dynamic data mining technique for rules extraction in a process of battery charging

Battery charging controllers design and application is a growing industry direction. Fast and efficient charging of battery packs is a problem which is difficult and often expensive to solve using conventional techniques. The majority of existing works on intelligent charging systems are based on expert knowledge and heuristics. Not all features of the desired charging behavior can be attained by the hard-wired logic implemented by expert generated rules. Because the battery charging is a highly dynamic process and the chemical technology a battery uses varies significantly for different battery types, data mining technique can be of real importance for extracting the charging rules from the large databases, especially when the charging logic is to be continuously changed during the life of the battery dependent on the type and characteristics of the battery and utilization conditions. In this paper we use soft computing-based data mining technique for extraction of control rules for effective and fast battery charging process. The obtained rules were used for NiCd battery charging. The comparative performance evaluation was done among the existing charging control methods and the proposed system, which demonstrated a significant increase of performance (minimum charging time and minimum overheating) using the soft computing-based approach.

Advances in analytics: Integrating dynamic data mining with simulation optimization

We introduce a simulation optimization approach that is effective in guiding the search for optimal values of input parameters to a simulation model. Our proposed approach, which includes enhanced data mining methodology and state-of-the-art optimization technology, is applicable to settings in which a large amount of data must be analyzed in order to discover relevant relationships. Our approach makes use of optimization technology not only for data mining but also for optimizing the underlying simulation model itself A market research application embodying agent-based simulation is used to illustrate our proposed approach.

Promoting performance and separation of concerns for data mining applications on the grid

Grid Computing brought the promise of making high-performance computing cheaper and more easily available than traditional supercomputing platforms. Such a promise was very well received by the data mining (DM) community, as DM applications typically process very large datasets and are thus very resource intensive. However, since the Grid is very dynamic and parallel data mining is prone to load unbalancing, obtaining good data mining performance on the Grid is hard. It typically requires the scheduler to understand the inner workings of the application, bringing two related problems. First, good Grid schedulers tend to be very specialized in the application they target. Second, changing the application may require changing the scheduler, which may be especially challenging when there is no clear separation between the application and the scheduler code. We here propose and evaluate a knowledge-based approach that provides abstractions to the DM developer and optimizes at runtime the DM application on the Grid.

Dynamic visual data mining: biological sequence analysis and annotation using SeqVISTA

In the post-genomic era, the volume of public sequence databases is increasing exponentially and visualisation-centric techniques have become more and more important in biological sequence analysis and annotation. In this paper, we present a methodology called dynamic visual data mining (DVDM), which combines biological object modelling, interactive display, and data analysis tools into one integrative platform. Using Java Development Kit v1.4, an object-oriented software named SeqVISTA has been developed based on DVDM. To illustrate the application of SeqVISTA, the following examples are shown: regular expression pattern matching; comparative analysis of alternative exon splicing patterns; Fourier analyses; exon prediction (MZEF and GENSCAN). Overall, we argue that DVDM is an important technique for biologists to unveil the information hidden behind the large genomic and proteomic databases, and SeqVISTA provides a versatile tool that integrates multiple computational algorithms for meeting biologists' data mining needs.

A methodology for dynamic data mining based on fuzzy clustering

Dynamic data mining is increasingly attracting attention from the respective research community. On the other hand, users of installed data mining systems are also interested in the related techniques and will be even more since most of these installations will need to be updated in the future. For each data mining technique used, we need different methodologies for dynamic data mining. In this paper, we present a methodology for dynamic data mining based on fuzzy clustering. Using the implementation of the proposed system we show its benefits in two application areas: customer segmentation and traffic management.

135時系列ボリュームデータマイニングのための位相解析アプローチ

This paper takes advantage of a 3D field topology analysis for automating transfer function design aiming at volume data mining. The conventional Reeb graph-based approach to describe the topological features of 3D surfaces is extended to capture the topological skeleton of a volumetric field. Based on the analysis results, which are represented in the form of hyper Reeb graph, we propose two principles to design appropriate color/opacity transfer functions for direct volume rendering. Feasibility study of the present methodology is performed with a large scale 4D simulated dataset from atomic collision research. In analyzing the dynamic volume data, it turned out to be effective to use transfer functions designed appropriately at each time step. The basic idea of another approach with a higher order hyper Reeb graph to dynamic volume data mining is also discussed.

Dynamic medical data fusion and data mining in the wound-healing process

Wound healing assessment is dependent on the quality of data collected about a patient and his or her wound and the clinician’s ability to evaluate that data and produce a diagnosis. Wound data has very little structure, with most patient-based data being stored in hard copy patient notes. The process of wound healing would benefit from a structured database to store all the different data types required and provide the clinician with a quick and efficient tool to display any data required. To achieve this, data visualization and data fusion will be used to bring together different data modalities in real time. The application will allow users to decide which data, or combination of data, they wish to view. Data mining will be applied to the wound data set, in an attempt to extract any hidden knowledge that is not apparent. The database structure itself will use a novel approach: object-oriented data streaming. This will allow new data types to be incorporated into the database, dynamically and transparently. By structuring the wound data a first step will be taken towards laying down a classification standard for wound-healing assessment.