جستجوی مقالات مرتبط با کلیدواژه « سیستم توصیه گر » در نشریات گروه « فنی و مهندسی »

Recently, recommender systems have expanded more and more as a new and fundamental technology to support users in choosing the right resources. These systems provide a personalized environment for selecting the desired resources by examining the past interactions of users and identifying interests. Of course, user behavior modeling and the recommendation mechanism are fundamental and decisive issues in the efficiency of recommender systems. In the field of e-commerce, the use of recommender systems plays an essential role in improving the user experience, attracting potential customers, increasing sales, and optimizing the efficiency of related service systems. Therefore, considering the importance of these systems in today's electronic businesses, knowing the functional dimensions of recommender systems is of particular importance. In this article, we are going to review the basic dimensions of recommender systems in the field of e-commerce and introduce some practical tools in this field. Certainly, by moving towards e-commerce systems based on intelligent recommender systems, we will witness huge and revolutionary changes in the infrastructure of the digital economy and related services.

Recommender systems are a subset of intelligent information filtering systems that discovers user interests and provide user-friendly recommendations. User-based collaborative filtering recommender systems is one of the most important types of recommender systems. However, they are faced with voluminous data and sparsity problems that have negative effects on the performance of the systems. In the proposed method, fuzzy C-ordered means clustering algorithm is integrated with a chaotic self-adaptive particle swarm evolutionary algorithm for clustering users. The proposed method aims to improve the rating prediction in large sparse datasets and reduce the negative impact of outliers and noisy data. Experiments have been conducted on real-world datasets to evaluate and prove the efficiency of the proposed method. Experimental results show the superiority of the proposed method that the state-of-the-art methods based on prediction error criteria, accuracy rates, and the computational time.

The Internet of Things is an emerging information architecture based on the Internet that develops interaction between things and services in a safe and reliable environment. In fact, the purpose of this structure is to reduce the distance between the things of the physical world and information systems. In the Internet of Things, it is expected that intelligent devices will become active members in business and informational and social processes, so that they are able to interact between themselves and the external environment through the exchange of data and sensed information. In fact, the Internet of Things is a network of devices in which various things can communicate with other equipment with the help of computers and through Internet connections.  Recommendation technologies can help to more easily identify relevant artifacts and thus will become one of the key technologies in future IoT solutions.  The main task of recommender systems is to recommend service providers that meet the different needs of users. The paper porposes a Support Vector Machine (SVM) based algorithm and the TOPSIS multi-criteria decision-making model in order to create an effective recommender system and provide suggestions to users based on their preferences and increase user satisfaction. The experimental results show that the proposed recommender system can produce a series of objective recommendations that are effective based on accuracy and variety, novelty and high coverage. Finally, the results confirm the improvement in making recommendations .

Recommender systems can predict future user requests and then generate a list of the user's favorite pages. In other words, recommender systems can obtain an accurate profile of users' behavior and predict the page that the user will choose in the next move, which can solve the problem of the cold start of the system and improve the quality of the search. In this research, a new method is presented in order to improve recommender systems in the field of the web, which uses the DBSCAN clustering algorithm to cluster data, and this algorithm obtained an efficiency score of 99%. Then, using the Page rank algorithm, the user's favorite pages are weighted. Then, using the SVM method, we categorize the data and give the user a combined recommender system to generate predictions, and finally, this recommender system will provide the user with a list of pages that may be of interest to the user. The evaluation of the results of the research indicated that the use of this proposed method can achieve a score of 95% in the recall section and a score of 99% in the accuracy section, which proves that this recommender system can reach more than 90%. It detects the user's intended pages correctly and solves the weaknesses of other previous systems to a large extent.

Recommender system can be defined as a software that suggests the most appropriate and closest item to the user's taste. They work as a counselor, behaving in such a way to guide people in the discovery of products of interest.Nowadays A great number of recommendation methods are used to implement a recommender system, a group of these algorithms are called collaborative filtering. These methods use the similarity between users or the similarity between items according to their user rating patterns for generating recommendations. Collaborative filtering algorithms can recommend the user, interesting items which are not similar to items she has rated before. These recommendations are generated according to the preferences of users with similar taste to the target user.  Different similarity functions and metrics have been used to create the model or compute the similarity in collaborative filtering methods. The best method which generates the most relevant items is not always the same and it may change according to the available data of users and items, because each approach has particularities and depends on the context to be applied. Thus, it becomes a hard task for system designers to manually select an appropriate method among the techniques.This article proposes an approach based on genetic algorithm for rank aggregation of memory based collaborative filtering methods and chooses the most relevant recommendations generated by different similarity techniques to create a Top-N recommender system. In order to implement these techniques, in addition to computing the similarity between users, inferred trust is also computed to increase the amount of available information about relations between user interests. The final method proposes a combination of collaborative filtering techniques for each data set, which in addition to considering time limits, has an acceptable precision for making recommendations.The proposed method has been compared against memory based collaborative filtering methods and similar methods. Experiments were performed using 1M MovieLens and 100k MovieLens and HetRec2011 data sets. The results show that the methodology proposed in this paper performs better and has a higher precision in generating recommendations for users than any of similar algorithms.

In this research, a new method is presented to improve the recommendation systems in the field of health tourism, which can make accurate predictions by using participatory filtering and by using the points that previous tourists have given to places and health professionals in our country. For the use of tourists. According to the research, data clustering using DBSCAN algorithm obtained 99% efficiency score, which is the highest efficiency score among the existing algorithms. Also, SVM method has 95% score in accuracy section and 99% score in call section. Which shows the high accuracy of predicting the results and the proposed method in general up to 80% can correctly identify the places needed by the tourist and suggest the appropriate place to a large extent correctly

In this research, a new method is presented to improve the recommendation systems in the field of health tourism, which can make accurate predictions by using participatory filtering and by using the points that previous tourists have given to places and health professionals in our country. For the use of tourists. According to the research, data clustering using DBSCAN algorithm obtained 99% efficiency score, which is the highest efficiency score among the existing algorithms. Also, SVM method has 95% score in accuracy section and 99% score in call section. Which shows the high accuracy of predicting the results and the proposed method in general up to 80% can correctly identify the places needed by the tourist and suggest the appropriate place to a large extent correctly

One of the significant dangers to human life is crises and natural disasters worldwide every year. If such incidents are unpredictable, their risks and casualties will be much higher. Among disasters are floods, hurricanes, volcanoes, earthquakes, tsunamis. An earthquake is an event that is more prevalent than other disasters and is almost unpredictable. Respond structure to crises and disasters is called crisis management which deals with all issues before, during, and after crises and disasters and leads to activities in the field of planning, preparedness, prevention, response, and reconstruction. One of the most basic and essential things that can reduce the casualties of various events is disaster relief and rescue, which respond to the crisis management structure. Also, the contextual information of the environment, rescuers, and activities created a context-aware recommending system that can facilitate the process of interaction with the environment. This study has checked the types of contexts, their relationship, and the structure of earthquake rescue in Iran, where there is a significant crisis due to geographical location and seismicity. The whole problem space consists of three parts to provide a meaningful definition of the concept of "context" in the deployment of relief and rescue teams. The rescuer is the main context. Relief and rescue team as the object's environment, which includes team information of rescuers, consists of team members' position, distance, physical condition, and activity compared to other rescuers in the group. The physical environment is a collection of injured people, buildings, and relief and rescue teams in a specific area. Contexts of rescuers and their relationships, teams, and the hypothetical earthquake were studied by studying related articles and books and interviewing experts in the field of research. The study of context-aware and optimization methods used for the actual structure of rescue teams are the innovations of this study. Contexts during relief and rescue include location, time, the extent of human and building injuries, rescuers' interactions with and with the environment, and activities, rescuers' specialties, and priorities. Interaction is necessary for optimal management of relief and rescue. To create a relationship between the various contexts and optimize the relief and rescue process by defining the mathematical function and using sensed information from contexts into the proposed optimized algorithm. Finally, the solution has been designed and implemented with an ant colony algorithm and geospatial information system to optimize the allocation of rescuers to the affected areas and the necessary activities in the part of Tehran. The use of combination context-aware and artificial intelligence algorithms for the subject of relief and rescue in earthquake crisis is new research that Led to a 1.79-fold improvement of the proposed solution compared to not considering the existing contexts in relief and rescue without using artificial intelligence algorithms. So can be created a context-aware system based on the appropriate optimization algorithm as a suitable solution to the problem of post-earthquake relief and rescue. Due to the context structure in this research in the individual's activity effectiveness on other individuals and groups, the ant colony algorithm is a collective intelligence base that can provide optimal positioning in a discrete environment. It allows more repetition in less time than other algorithms.

Nowadays recommender systems have become an integral part of e-commerce websites. However the general and accessibility of these systems has made them vulnerable to attack by profiteering users. Numerous studies have examined the vulnerability of various recommender algorithms to attacks created by creating fake profiles, many of which have focused on older methods, including group refinement algorithms.A group of recommender algorithms considered by various Internet services use a variety of graph analysis methods, including random steps, to provide suggestions to the user. There are limited studies on the vulnerability of graph-based recommender algorithms that focus on specific types of these methods. Therefore, in this paper, the vulnerability of a group of graph-based algorithms that use the idea of ​​PageRank ranking algorithm on the web to score items and generate their suggestions was examined.To do this, a new attack model called the link farm is proposed using the PageRank ranking algorithm applied. The results obtained from the application of different attacks to these techniques have shown that the proposed attack model affects this group of graph-based recommender algorithms.

In this research, a new method is presented to improve the recommendation systems in the field of health tourism, which can make accurate predictions by using participatory filtering and by using the points that previous tourists have given to places and health professionals in our country. For the use of tourists. According to the research, data clustering using DBSCAN algorithm obtained 99% efficiency score, which is the highest efficiency score among the existing algorithms. Also, SVM method has 95% score in accuracy section and 99% score in call section. Which shows the high accuracy of predicting the results and the proposed method in general up to 80% can correctly identify the places needed by the tourist and suggest the appropriate place to a large extent correctly

With the increasing popularity of sharing media on social networks and facilitating access to location technologies, such as Global Positioning System (GPS), people are more interested to share their own photos and videos. The world wide web users are no longer the sole consumer but they are producers of information also, hence a wealth of information are available on web 2.0 applications. The shared media usually contain geo-tagged locations, time stamp, hashtags, and comments. As such, mining social networks can yield extensive knowledge about human dynamics and mobility behaviors within urban context So web users are no longer just users but also producers of information This wealth of information can be leveraged for location-based services. If the locations visited by users are collected and sorted according to the timestamps, the sequence that the user has visited can be determined; exploring of which can be used in tourism planning. Recently, there is an increasing tendency to adopt the information from these geo-tagged photos for learning to recommend tourist locations. For a tourist, before traveling to an unfamiliar city, the most important preparation is planning the trip. without any prior knowledge, tourist must either rely on travel books, personal travel blogs or combination of online resources and services. It is difficult and time consuming and painstaking to find out the locations worth to visit and figure out the order in which they are to be visited. Hence, the purpose of the present study is to provide a framework for recommending locations and travel sequences to tourists by using geo-tagged photos in social networks. Most existing methods for tourist recommendation do not consider context constraints, or at best, address a few dimensions of contexts. The present work aims to develop a context-aware recommender system that recommends interesting locations and the travel sequence. The proposed method is designed such that it can use the collective wisdom of people from collection of geo-tagged photos in order to provide a set of tourism locations and interesting trip sequences that matches the user's current context given a city that is unfamiliar to that user.  first Due to the low accuracy of positioning with GPS embedded in mobile phones to find a unique pair of geographic coordinates for a tourist place the geo-tagged photos were clustered. For this reason OPTICS clustering method exploit to group geo-tagged photos by their locations. It then uses the combined method, to annotate all of the clusters that are created in the previous step with semantics. Then, we create a profile for clusters by using historical context (time stamps and weather). After that, we generated a travel sequences database and rated the sequences in the database according to their context. Finally In order to evaluate the performance of the proposed method, Panoromia site dataset of one region in Tehran was used and Experimental results showed that 64.5% of the results obtained by our proposed strategy are identical with the user preferences, which illustrate rationality of the recommendation from analyzing the geo-tagged photos.

Nowadays, tourists are planning trips by their own using the available services on the web. Travelling individually does not have the pleasure of group tours, so some tourists decide to find similar friends on their destination to share the joy of tours and trips. Also, they may not be familiar with the routes available in the city. In this paper, a mobile application is designed, developed, implemented and evaluated to recommend similar tourists to each other and generate optimal route based for each individual. This system is integrating mobile GIS, recommender systems and artificial intelligence tour planning algorithms into a single application. Being on a mobile platform is necessary, especially for navigation and tour planning systems. User may have to carry his cellphone in order to follow the suggested tour. The recommender system is a demographic filtering recommender to find the similar tourists. It considers age, nationality, education and interest to find the proper matches. After suggesting similar tourists to each other, one of the versions of Ant Colony Optimization namely Ant System is used to plan the optimal tours for each individual. This problem can be defined as a new version of Multiple Travelling Salesman Problem with mutual nodes to visit simultaneously. Different tourists decide to share a tour that stay in different places. This system should find a proper sequence of sightsees to minimize the length of the tour and also it should maximize the number of mutual streets in a tour to let tourists travel together during the tour. The objective function is to minimize the sum of the distances each tourist traversed and to maximize the number of mutual streets with a proper POI sequence. Maximizing number of mutual streets was considered because tourists want to travel as a group in a shared trip. Implementation of the system took place in Shiraz, Iran. Shiraz is the fifth-most-populous city of Iran and the capital of Fars Province with more than 4000 years of history that made it one of the key tourism sites in Iran. Numbers of questionnaires were distributed among different people to evaluate the people recommendation system. In addition, tour planning algorithm was evaluated using a well-known evolutionary algorithm (i.e. Genetic Algorithm). Mean, standard deviation and the convergence graphs of the two mentioned algorithms were compared. Results indicated accurate performance of the recommender system and high accuracy and precision of the route planning algorithm. Recommender System had a mean difference of 0.2 with the questionnaire results, which indicated its good functionality. Ant System reached the minimum value of the objective function (34.70) with a better standard deviation (0.61) compared to Genetic Algorithm. However, Genetic Algorithm performed better in mean value of the tests (34.32) which is a measure of the precision. Convergence graph of the Genetic Algorithm showed a fast convergence with lower objective values in the beginning. Ant System convergence graph showed a smooth convergence toward the optimal solution with an initial population with higher objective values. This indicated the proper functionality of the Ant System and the possibility of improving the results with generating better initial population.

As the number of published geospatial resources in the web is increasing permanently, providing users with proper tools for search and discovery of these resource is of great importance. Geoportals are a type of web portals that enable users to find geospatial resources. However, when users search for a specific resource in existing geoportals, a wide range of unrelated results might be retrieved. Such results are often confusing, so that user has to spend a lot of time to find the most appropriate resource. In addition, working based on user desires and preferences has been considered as a fundamental characteristic of modern web applications.
In this article, the capability of recommendation is added to the geoportal. The recommendation capability enhances the ability of geoportals which is able to recommend the most appropriate geospatial resources to users by considering their desires and preferences. In order to accomplish this purpose, a new solution with recommender systems is proposed. Recommender systems using knowledge derived from the users’ previous interactions can lead users to access resources that they are interested in. Collaborative filtering is a common technique used in recommender systems that deal with this problem. This method uses user rating data to extract the similarity between users or resources for making recommendations. Moreover, mathematical functions are defined in order to improve the efficiency of the recommender geoportal process. The functions are designed according to the specifications of users and resources such as city, county and language, along with the distance between the user and resource.
In addition, an operation of these functions is to obviate the cold-start problem in collaborative filtering. To evaluate the designed geoportal recommender, a recommender geoportal is implemented so that a user is able to profit the advantages and usage of its recommendations. The obtained results indicate that the efficiency of recommender geoportals is improved compared with common geoportals.

The rapid growth of e-commerce has led to introducing various products from companies which are active in this area. As a result, choosing an appropriate product has become difficult for a customer. Recommender systems aimed at offering the best and most appropriate items to users based on their interests and their previous activities. This paper propose a hybrid recommender system which combines two approaches: collaborative filtering and constraint satisfaction. The proposed system, on the one hand, solves the cold start problem of the collaborative filtering technique because of using constraints. On the other hand, it solves the problem of constructing the knowledge base, which is needed for constraint-based methods. Instead of constructing the knowledge base manually, which is time consuming, the knowledge base of the constraint-based approach is extracted from interactions among the target user’s neighbors. Using the "Movielens" dataset for evaluation, the experimental results show that the performance of the proposed approach is better than the collaborative filtering method.