CONCLUSION

In summary, this research work investigated the OntoQuery system within an m-commerce agent framework against current query formation and information retrieval systems.

The prototype implementation results showed that querying formation using an ontology approach is efficient, as it provides a friendly environment to the user . In addition, by combining the keyword and ontology approaches, a more efficient and effective way of forming queries could be achieved. Thus, the objective to propose efficient query formation for product databases is successful.

It was found that the genetic algorithm is able to optimize queries effectively. Also, using genetic approaches, we have proposed and tested out various fitness functions for searching product databases. The use of correlation prevents the original query from diverging. In addition, restructuring of the logical terms and numerical constraints in queries served as an effective way of constraint relaxation for mutated queries to respond to the situation of no documents retrieved. Moreover, adding feedback to the system helps it to cater to the needs of the user more closely. It also helps to maintain a converging query trend.

It was also found that the efficiency of our genetic algorithm initially increases, reaches a maximum value, and then decreases as population size increases . Thus, there is an optimal value of population size when the genetic algorithm is applied in query optimization.