Developing a Hybrid Data Mining Approach Based on Multi-Objective Particle Swarm Optimization for Solving a Traveling Salesman Problem

Abstract

A traveling salesman problem (TSP) is an NP-hard optimization problem. So it is necessary to use intelligent and heuristic methods to solve such a hard problem in a less computational time. This paper proposes a novel hybrid approach, which is a data mining (DM) based on multi-objective particle swarm optimization (MOPSO), called intelligent MOPSO (IMOPSO). The first step of the proposed IMOPSO is to find efficient solutions by applying the MOPSO approach. Then, the GRI (Generalized Rule Induction) algorithm, which is a powerful association rule mining, is used for extracting rules from efficient solutions of the MOPSO approach. Afterwards, the extracted rules are applied to improve solutions of the MOPSO for large-sized problems. Our proposed approach (IMOPSP) conforms to a standard data mining framework is called CRISP-DM and is performed on five standard problems with bi-objectives. The associated results of this approach are compared with the results obtained by the MOPSO approach. The results show the superiority of the proposed IMOPSO to obtain more and better solutions in comparison to the MOPSO approach.

A profile‐boosted research analytics framework to recommend journals for manuscripts

“…We follow the Multi Objective Particle Swarm Optimization (MOPSO) as in Haeri and Tavakkoli‐Moghaddam () to solve our multiobjective journal recommendation model presented above, for it can generate two main advantages over the others. First, it is a very simple approach when compared to the other traditional optimization algorithms such as genetic algorithms (GA), as it considers only one operator for creating a new solution (Reyes‐Sierra & Coello, ) leading to less computational complexity.…”

“…Based on the evaluation results the nondominated solutions from the swarm are selected. A solution j is said to be nondominated, if there are no other solutions that exceed the objective values of solution j (Haeri & Tavakkoli‐Moghaddam, ).…”

“…The solutions with lower CD values are excluded from the list. This implies that solutions which are more crowded with other solutions are removed to maintain more diversification in the space search process when executing the algorithm (Haeri & Tavakkoli‐Moghaddam, ). This guarantees that the final recommended list of journals consists of a diversified set rather localizing only to a very similar set of journals.…”

“…With each iteration, the velocities of the particles are updated. Following the equation presented in Haeri and Tavakkoli‐Moghaddam () we update the velocity of each particle as follows: where v i + 1 and v i represent the velocity vectors in the i + 1 th and i th iteration, respectively, pbest i is the best position of the particle of its i th iteration, x i is the position vector in the i th iteration. c 1 and c 2 represent predefined coefficients as r 1 and r 2 represent random numbers in [0,1].…”

“…w is the inertia factor that can be equal to 1. rep H is the position vector of the representative solution H which has the lowest CD, selected from the solution repository. We follow the equation presented in Haeri and Tavakkoli‐Moghaddam () to update the particle positions as follows. …”