|
In the Internet age, almost all kinds of services and products are available online for selection and use. In addition, there are several different providers for one type of product or service. In this context, to compare and find the right service according to the end customer, a service recommendation system is required. The purpose of the recommendation system is to understand the current requirements of the client and to study the database to restore the most likely services. To demonstrate the problems of this area and their solutions, a real problem is used, namely food and grocery delivery services. According to the design task, this recommender system is treated as a search system for a structured data source. Thus, the quantum genetic method is used to find suitable results from the proposed working model. This method first takes information from the data set and the user's requirement and then encodes the information in binary values. Next, the request sequence is processed as a binary string with all ones. Finally, a genetic algorithm is implemented to find a suitable solution among all available binary sequences. The generated results from the genetic algorithm are considered as the final recommendation of the search engine. In addition, fitness values are used to rank solutions. Implementation and evaluation of results are performed in C#. After that, performance is measured using time and space complexity. The obtained performance parameters demonstrate the acceptability of the work.
Keywords:data mining, recommender systems, genetic algorithm, search, databases
|