مقاله انگلیسی رایگان در مورد معرفی نام دامنه بر اساس شبکه عصبی – الزویر 2018

1- Introduction

2- State of the art

3- The proposed approaches

4- Experimental results

5- Conclusion

References

Abstract

The number of website is increasing speedily, and clients purchase their website from the enterprise that suggests them the best domain name with a good price. In order to give the relevant domain name, enterprise is always eager to have a good system of suggestion that suits the client request. Recommender system has been an effective key solution to guide users in a personalized way for discovering the domain name they might be interested in from a large space of possible suggestion. They have become fundamental applications that provides to users the best domain name that meet their needs and preferences. In this work, we used a recommender system based on neural network as it is capable to solve many complex tasks and gives better customer satisfaction. We proposed two approaches to recommend domain name, both of these approaches are based on neural network. The first one consists on discovering the similarity between the vocabulary of domain name, while the second one is finding the relevant Top Level Domain (TLD) corresponding to the context of the domain name. First experiments on GANDI datasets shows the effectiveness of the proposed approaches.

Introduction

Recommender systems can be defined as programs which attempt to recommend the most suitable items, products or services to particular users by predicting a user’s interest in an item based on related information about the items, the users and the interactions between items and users. The aim of developing recommender systems is to reduce information overload by retrieving the most relevant information and services from a huge amount of data, thereby providing personalized services. The most important feature of a recommender system is its ability to guess a user’s preferences and interests by analyzing the behavior of this user or the behavior of other users to generate personalized recommendations. Various recommender system techniques have been proposed, and many sorts of recommender system software have been developed recently for a variety of applications. Researchers recognize that recommender systems offer great opportunities and challenges for business, government, education, and other domains, with more recent successful developments of recommender systems for real-world applications becoming apparent. The past few decades have witnessed the tremendous successes of the use of neural network in many application domains such as speech recognition, image analysis and natural language processing. Applying neural network into recommender system has been gaining momentum thanks to its state-of-the-art performances and high-quality recommendations. In contrast to traditional recommendation models, neural network provides a better understanding of user’s demands, item’s characteristics and historical interactions between them. In this paper, we suggest a system of recommendation of fully qualified domain name based on neural network. When a customer wants to purchase a domain name he rarely knows his qualified domain name, therefore, we will make a system of recommendation to suggest to the customer both the domain name and the extension that suits his recommend by referring to his localization and his research. The first part is the state of the art where we will present the existing methods concerning text similarity, through three approaches: Probabilistic-based approach, Predictionbased approach and Hybrid approach which is the combination of the two previous approaches. The second part is the description of the two models for suggesting domain name that are based on neural network: the first method tries to find similarity based on neighbors, while the second one predicts the TLD (Top Level Domain) corresponding to the context of domain name. The third part concerns the experiment results for the two models based on the real purchase database of GANDI.