Given a problem, the human brain has the capability of identifying relationships in data, learn from them and create solutions. However, the new technologies have allowed generating and increasing huge amount of information and, consequently, experts are not able to process them in the same way than they did some years ago. In this sense, Decision Support System has become a great opportunity in industry thanks to their capabilities of helping experts to take better decisions because they are able to manage this new scenario.
The development of intelligent systems for helping experts to give support for their decisions is based on (1) extracting interesting patterns from moderate and large complex data (Data Mining) and (2) learning from them (Machine Learning). To do it, several approaches based on Soft-Computing techniques are used such as Evolutionary Computation, Case-Based Reasoning, Neural Networks, Complexity Metrics and Fuzzy Logic. You can find more information in the publication section about these topics.
Figure 1. Decision support system solves new problems using knowledge and patterns extracted from the domain when data is too complex and large and expert is not able to do it without support.
Figure 1 describes a general Decision support system framework characterized by having a knowledge base composed by knowledge extracted from the expert and also for patterns extracted from the problem domain. Given a new problem, system looks for useful knowledge inside the knowledge base for reusing it in order to propose a new solution. Once upon this solution has been proposed, it is showed to the expert in way that expert can understand why this solution is interesting and expert can assess the utility and the quality of the solution. Thus, this feedback is used to update the knowledge base and improve the performance for the next times.
Decision support systems can be used for a wide range of applications such as for example:
• Searching for relationships between variables
• Building a model generalizing for classification task
• Find a function which models the data with the least error
• Grouping information according to a set of criteria
We have successful applied this approach in industry, highlighting some projects such as for example::
• Decision support system for breast cancer diagnosis using the knowledge obtained from the analysis and diagnosis of previous patients
• Melanoma cancer characterization using the knowledge obtained from the analysis and diagnosis of previous patients
• Managing more efficient the demand of energy by the identification of groups of consumers with similar behavior using the historic of electrical consumption.
• Integration and management of communication technologies in smart-grids for assuring QoS, security and reliability using automatic decision system.
• Detection of security vulnerability in networks by means of the identification of patterns.
• Forecasting of risk premium in car insurances based on driving habits and the places where he goes.
• Recommender systems in a Marketplace using constraint satisfaction problems and user experience.
Decision Suport SystemAlbert Fornells email@example.com