Journal of Al-Qadisiyah for Computer Science and Mathematics

Any phenomenon may be considered by a set of variables that difficult to treat with and explain it at first time without reduce its variables into lower number which is the hope of any researcher deal with principle component analysis and factor analysis.This paper tries to applay factor analysis technique by using principle component method on actual data and many conclusions and recommendations are reached

We apply the no-arbitrage and no-free lunch definitions of Kreps (1981). Arbitrage is a linear algebraic notion, while free lunch is a topological notion. The notion of free lunch, unlike that of arbitrage, is somewhat ambiguous because it depends on a topological choice that is often tacit, i.e. there is source of problem when we use the topology in our work.
Our main point in this paper is that how the choice of a topology which the space are free lunches. We consider a number of topological spaces depending on:
(1) not a free lunch. (does not converge).
(2) a free lunch (converge to a positive limit).
(3) not a free lunch (converge to a non-positive limit).

This work consists of two sections. In section one we recall the definition concerning semi-open sets. In section two we recall the definition of paracompact space and we introduce similar definition by using semi open sets also we proof some results about it.

In this paper a new method is proposed for the solution of special case of two-point second order boundary-value problems( TPBVP ) ,that is we concerned with constructing polynomial solutions to two point second order boundary value problems for ordinary differential equation.
A semi-analytic technique using two-point osculatory interpolation with the fit equal numbers of derivatives at the end points of an interval [0,1] is compared with conventional methods via a series of examples and is shown to be that seems to converge faster and more accurately than the conventional methods and generally superior, particularly for Variations on a problems .
Also we introduce some general observations about control of a residual and control of the true error and we prove, there is a more useful connection between scaled residual and true error.

We define an paraLindelöf, countable paraLindelöf, semipara- Lindelöf, countable semiparaLindelöf topological and study some properties of these concepts and give the relation between these concepts. And we give the relationship between the paraLindelöf space and regular (normal) space.

In this paper, the concepts of submodule with finite submodule-length, and module with finite submodule-lengths are introduced. These concepts are generalizations of the concepts of ideal with finite ideal-length and ring with finite ideal-length.
A submodule N of an R-module M is said to have finite submodule length if comp(I) is finite and (M / I M)P , is an Artinian and Noetherian (R / I)P-module,  P = P / I , I = (N:M) and P  comp(I). An R-module M has finite submodule-lengths if each submodule of M has finite submodule-length. The basic results about these concepts, and some relationships between modules with finite submodule-lengths and other classes of modules are given.

Let be an BA- finitly additive set function on a measurable space . In this paper, we proved the following statement
(1) If is BA- countably additive, then is continuous at for all .
(2) If is - continuous from below at every set , then is BA- countably additive .
(3) If is , finitly and continuous from above at , then is BA- countably additive.

In this study, to minimize a function of two cost criteria for scheduling n jobs on a single machine , the problem is discussed :
“ Minimizing a function of total square completion time and maximum Earliness simultaneously”.
For this problem we proposed some algorithms to find exact(optimal) solution for hierarchical case and efficient (pareto optimal) solutions for simultaneous case. Also we proposed branch and bound algorithm to find exact solution for sum of total square completion time and maximum Earliness ,and present algorithm D to find exact solution in a fast way with respect to (BAB) method. We present computational experience for the (BAB) method and algorithm(D) on a large set of test problems

Genetic Programming is one of the evolutionary algorithms developed to solve wide area of industrial and scientific problems. Rather than dealing with population of string like Genetic Algorithm, Genetic Programming composes the first population from programs tree derived from the function set of the problem. In this paper, we extend the selection algorithm of the GP by using the Learning Classifier System, which build and derive the Hypothesis set from the population. The selection algorithm redesigned to enforce the selection been added from the Hypothesis domain. The proposed system called ILPSAGP was built using C#.net 2008 and tested with traditional problem like Line Regression problem. The obtained results shows more accurate result than traditional Genetic Programming.

A computational investigation is carried out in the field of charge particle optics with the aid of the numerical analysis method using personal computer. The work concerned with the design of three electrode unipotential electrostatic lens with low spherical and chromatic aberration .

The axial electrostatic potential distribution is determined by using polynomial functions of the third and forth order from which the paraxial ray equation is solved to obtain the trajectory of particles that satisfy the suggested potential function.

From the knowledge of the first and second derivatives of the axial potential distribution the spherical and chromatic aberration are determined, the electrode shape of electrostatic lens was then determined from the solution of the paraxial ray equation.

Let R be a 2-torsion free *-ring, and d: R→R be a Jordan *-derivation. In this paper we prove the following results: (1) If R is a non-commutative prime *-ring, and d(h) h + h d(h)  Z(R) for all h  H(R), then d(h) =0 for all h  H(R).(2) If R be a non-commutative prime *-ring, and d([x,y])= [x,y] for all x, y  R, then R is normal *-ring.(3) If R is a semiprime *-ring, then there is no d satisfies d(xy+yx)=xy+yx for all x, y R, where H(R)={x; x R s.t x*=x }.