Search

The purpose of this system is to satisfy the expectations and requirements of the customers of computer assembly, which is a procedure that consists on correctly placing all the parts of a computer in order such that everything works efficiently. Then, it will be able to install the operating system and other programs according to the user's needs, thus a computer selection is based on information on compatibility and assembly of computers. The information has been compiled from books, scientific papers, and based on research tools such as: surveys, interviews, for the collection of information. The development of this system serves as a tool that in the future may bring some comfort to users with little knowledge of hardware to make purchases according to the needs and customer's economy, which is an important aspect in people's daily lives. The Expert System of requirements and compatibility in assembly of computers realizes what an expert engineer in assembly of computers can do, but the system does it in a fast way and it is always available. [ABSTRACT FROM AUTHOR]

Optimization based on Ant Colony two stages (TS -ac), it represents a way to improve exploration strategy meta-heuristic optimization based on Ant Colony (ACO). TS-ACO, it is based on the methodology of "divide and conquer" and its effectiveness has been proven in several discrete optimization problems. In this paper, three approaches to TS-ACO are presented and appropriate adjustments are made to test their effectiveness in solving the problem of Coverage Sets (SCP). These approaches use partial results (only sub solution (S-TS-ACO) or only pheromone trail (P-TS-ACO) or both (PS-TS-ACO)), obtained in the first phase, in order to guide exploration in the second phase of the search process. Experimental results show that the best solutions are achieved by the variant P-TS-ACO. Finally a comparative study of the two-stage strategy and the original ACO strategy was performed. The results show that the proposed P-TS-ACO, you get significantly better results than the original strategy, in the same execution time and under the same parameter values. [ABSTRACT FROM AUTHOR]

This paper proposes a comprehensive evaluation of supervised classification algorithms without semantic information for opinion mining on Twitter. The evaluation will assess the potential of this approach to tackle the given problem. Also, as result of evaluating, the impact of different variants of the text preprocessing is assessed, as well as different alternatives of dimension reduction techniques in the problem solution. Finally, the results obtained in the experiments offer evidence to direct the use of supervised classification to classify opinions and its valence in a domain of short messages such as Twitter. [ABSTRACT FROM AUTHOR]