**5. Conclusion**

136 Bio-Inspired Computational Algorithms and Their Applications

*Fitness Effectiveness Convergence. Average* 

*(18 hits)* 

*(46 hits)* 

*(65 hits)* 

*(93 hits)* 

Table 7. Comparative results Evolutionary System with various samples of documents showing the best results and the average results of evaluations with the "Distribution 21" of

Fig. 9. Graphs compare the results obtained with the composite function against Kmeans

Finally, to corroborate the results,we compare their results with the other collection in Spanish, which was processed in the same way, using all values of table 2. (see figure 10).

*Best Average* 

*Deviation Fitness* 

1163 0,15206069 0,00001601 *1165 17,8* 

*2079* 0,15304887 0,00004569 *2736 45,6* 

*2787* 0,15637693 0,00025014 *2810 66,4* 

*3359* 0,15720766 0,00024132 *1980 98,6* 

*Average* 

*Convergence Kmeans* 

*Fitness* 

*Distribution 21 Reuters* 

*Very Few documents (20 documents)* 

*Few Documents (50 documents)* 

*Many Documents (80 documents)* 

*Enough Documents (132 documents)* 

the Reuters 21578 collection

(four collection Reuters)

*Collection 4 Samples of documents* 

*Documents* 

*Best Result* 

*Categoríes: Acq y Earn* 

0,152048900 *90%*

0,153006650 *92%*

0,156029510 *81%*

0,157012180 *70,4%*

To then display the results graphically in figure 9.

In this study, we have proposed a new taxonomy of parameters of GA numerical and structural, and examine the effects of numerical parameters of the performance of the algorithm in GA based simulation optimization application by the use of a test clustering problem. We start with the characteristics of the problem domain.

The main characteristic features of our problem domain are:


These properties of the problem domain generate a rapid convergent behavior of GA. According to our computational results lower mutation rates give better performance. GA mechanism creates a lock-n effect in the search space, hence lower mutation rates decreases the risk of premature convergence and provides diversification in the search space in this particular problem domain. Due to the dominance crossover operator does not have significant impact on the performance of GA. Moreover, starting with a seeded population generates more efficient results.

We can conclude that the GA had a favourable evolution, offering optimal document cluster in an acceptable and robust manner, based on a proper adjust of the parameters. We proved that *the medium effectiveness* of the GA is very acceptable, being in most cases better than Kmeans supervised algorithm, but with the added advantage that we processed the documents in an unsupervised way, allowing evolution perform clustering with our adjustment. As a result of our experiments, we appreciate that we got the best performance with a rate of 0.03 for the mutation operator and using a rate of 0.80 for the crossover operator, this values appears to be ideal if we maximize the efficiency of the genetic algorithm.

**8** 

*México* 

*Autonomous University of Sinaloa* 

**Public Portfolio Selection Combining Genetic** 

**Algorithms and Mathematical Decision Analysis** 

Eduardo Fernández-González, Inés Vega-López and Jorge Navarro-Castillo

A central and frequently contentious issue in public policy analysis is the allocation of funds to competing projects. Public resources for financing social projects are particularly scarce. Very often, the cumulative budget being requested ostensibly overwhelms what can be granted. Moreover, strategic, political and ideological criteria pervade the administrative decisions on such assignments (Peterson, 2005). To satisfy these normative criteria, that underlie either prevalent public policies or governmental ideology, it is obviously convenient both to prioritize projects and to construct project-portfolios according to rational principles (e.g., maximizing social benefits). Fernandez et al. (2009a) assert that

• They may be undoubtedly profitable, but their benefits are indirect, perhaps only long-

• Aside from their potential economic contributions to social welfare, there are intangible benefits that should be considered to achieve an integral view of their social impact. • Equity, regarding the magnitude of the projects' impact, as well as the social conditions

Admittedly, the main difficulty for characterizing the "best public project portfolio" is finding a mechanism to appropriately define, evaluate, and compare social returns. Regardless of the varying definitions of the concept of social return, we can assert the

**Proposition 1**: *Given two social projects, A and B, with similar costs and budgets, A should be* 

Ignoring, for a moment, the difficulties for defining the social return of a project portfolio, given two portfolios, C and D, with equivalent budgets, C should be preferred to D if and only if C has a better social return. Thus, the problem of searching for the best projectportfolio can be reduced to finding a method for assessing social-project returns, or at least a

The most commonly used method to examine the efficiency impacts of public policies is "cost-benefit" analysis (e.g. Boardman, 1996). Under this approach, the assumed consequences of a project are "translated" into equivalent monetary units where positive

**1. Introduction** 

public projects may be characterized as follows.

tautological value of the following proposition.

*preferred to B if A has a better social return.*

of the benefited individuals, must also be considered.

comparative way to analyze alternative portfolio proposals.

term visible, and hard to quantify.

As a future research direction, the same analyses can be carried out for different problem domains, and with different structural parameter settings, and even the interaction between the numerical and structural parameters could be investigated.

#### **6. References**

