Hybrid Recommendation Algorithm Research Articles

Hybrid location-centric e-Commerce recommendation model using dynamic behavioral traits of customer

Major e-Commerce service provider offers additional product recommendation to its customers, while they access the application, and enough evidence existing that such recommendations are cost effective for both consumer and service provider. For maximizing profit and to satisfy the user, existing e-Commerce platforms use long-term context for recommendations. In actual scenario, the recommendation can aid the user for other reason such as when the product is reminded of recent interest in or, point customer to currently discounted items. Furthermore, user preference changes over time due to weather, location, etc. As a result, the recommendation must be made based on the present behavior of the ongoing session. Many research based on location and session-based approaches has been presented to forecast user’s next-item requirement. However, these models are not efficient, as they are designed either to model short-term or long-term preferences. Recently, some hybrid recommendation algorithms have been presented to model both short-term and long term, but these models are designed considering static behavior and finds difficulty in revealing the correlations among behaviors and items. Furthermore, these models do not consider location-centric information for performing recommendation. To overcome the above-mentioned challenges, our research work presents hybrid location-centric prediction (HLCP) model by considering the dynamic behavior traits of users. HLCP model can learn both short-term and long-term context efficiently. Experiment results show that HLCP attains significant performance over existing models in terms of mean reciprocal rate (MMR) and hit rate (HR).

Inferring microRNA-disease association by hybrid recommendation algorithm and unbalanced bi-random walk on heterogeneous network

More and more research works have indicated that microRNAs (miRNAs) play indispensable roles in exploring the pathogenesis of diseases. Detecting miRNA-disease associations by experimental techniques in biology is expensive and time-consuming. Hence, it is important to propose reliable and accurate computational methods to exploring potential miRNAs related diseases. In our work, we develop a novel method (BRWHNHA) to uncover potential miRNAs associated with diseases based on hybrid recommendation algorithm and unbalanced bi-random walk. We first integrate the Gaussian interaction profile kernel similarity into the miRNA functional similarity network and the disease semantic similarity network. Then we calculate the transition probability matrix of bipartite network by using hybrid recommendation algorithm. Finally, we adopt unbalanced bi-random walk on the heterogeneous network to infer undiscovered miRNA-disease relationships. We tested BRWHNHA on 22 diseases based on five-fold cross-validation and achieves reliable performance with average AUC of 0.857, which an area under the ROC curve ranging from 0.807 to 0.924. As a result, BRWHNHA significantly improves the performance of inferring potential miRNA-disease association compared with previous methods. Moreover, the case studies on lung neoplasms and prostate neoplasms also illustrate that BRWHNHA is superior to previous prediction methods and is more advantageous in exploring potential miRNAs related diseases. All source codes can be downloaded from https://github.com/myl446/BRWHNHA.

College Library Personalized Recommendation System Based on Hybrid Recommendation Algorithm

When the number of books provided by library is relatively large, it becomes difficult for user to select appropriate book from a lot of candidate books. In this case, this paper designs a personalized recommendation system for college libraries based on hybrid recommendation algorithm. First of all, paper studies the application of collaborative filtering and content-based recommendation algorithm in the recommendation of university books, which involves reader classification, the establishment of user-item scoring matrix, the construction of vector space model and the calculation of similarity among users. And considering the characteristics of books and readers in universities, the user - item scoring matrix is improved, and clustering is used to alleviate the data sparsity problem. Do comparative experiments using the hybrid algorithm in data sets of Library of Inner Mongolia University of Technology. The results demonstrate that the hybrid methods can provide more accurate recommendations than pure approaches. Finally, the Spark big data platform combined with the hybrid recommendation algorithm is used to achieve the personalized book recommendation system design.

A Medical-History-Based Potential Disease Prediction Algorithm

As an important application of medical informatization, healthcare big data analysis has been extensively researched in the fields of intelligent consultation, disease diagnosis, intelligent question-answering doctors, and medical assistant decision support, and have made many achievements. In order to improve the comprehensiveness and pertinence of the medical examination, this paper intends to use healthcare big data analysis combined with deep learning technology to provide patients with potential diseases which is usually neglected for lacking of professional knowledge, so that patients can do targeted medical examinations to prevent health condition from getting worse. Inspired by the existing recommendation methods, this paper proposes a novel deep-learning-based hybrid recommendation algorithm, which is called medical-history-based potential disease prediction algorithm. The algorithm predicts the patient's possible disease based on the patient's medical history, providing a reference to patients and doctors to reduce the problem of delaying treatment due to unclear description of the symptom or limited professional knowledge. The experimental results show that our approach improves the accuracy of the potential diseases prediction.

基于机器学习的融合推荐算法研究

基于机器学习的融合推荐算法研究

On Recommendation of Learning Objects Using Felder-Silverman Learning Style Model

The e-learning recommender system in learning institutions is increasingly becoming the preferred mode of delivery, as it enables learning anytime, anywhere. However, delivering personalised course learning objects based on student's preferences is still a challenge. Current mainstream recommendation algorithms, such as the Collaborative Filtering (CF) and Content-Based Filtering (CBF), deal with only two types of entities, namely users and items with their ratings. However, these methods do not pay attention to student's preferences, such as learning styles, which are especially important for the accuracy of course learning objects prediction or recommendation. Moreover, several recommendation techniques experience cold-start and rating sparsity problems. To address the challenge of improving the quality of recommender systems, in this paper a novel recommender algorithm for machine learning is proposed, which combines students actual rating with their learning styles to recommend personalised course learning objects (LOs). Various recommendation techniques are considered in an experimental study investigating the best technique to use in predicting student ratings for e-learning recommender systems. We use the Felder-Silverman Learning Styles Model (FSLSM) to represent both the student learning styles and the learning object profiles. The predicted ratings are compared with the actual student ratings to determine the accuracy of the recommendation techniques, using the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) metrics. This approach has been experimented on 80 students for an online course created in the MOODLE Learning Management System. The results of the experiment show that the best recommendation technique is our proposed hybrid recommendation algorithm that combines the collaborative filtering and the content-based filtering techniques to enhance the accuracy of the predictions, and solves the cold-start and the rating sparsity problems using the FSLSM representations of the student learning styles and the learning object profiles.

Opening the Black Box: Explaining the Process of Basing a Health Recommender System on the I-Change Behavioral Change Model

Recommender systems are gaining traction in healthcare because they can tailor recommendations based on users' feedback concerning their appreciation of previous health-related messages. However, recommender systems are often not grounded in behavioral change theories, which may further increase the effectiveness of their recommendations. This paper's objective is to describe principles for designing and developing a health recommender system grounded in the I-Change behavioral change model that shall be implemented through a mobile app for a smoking cessation support clinical trial. We built upon an existing smoking cessation health recommender system that delivered motivational messages through a mobile app. A group of experts assessed how the system may be improved to address the behavioral change determinants of the I-Change behavioral change model. The resulting system features a hybrid recommender algorithm for computer tailoring smoking cessation messages. A total of 331 different motivational messages were designed using 10 health communication methods. The algorithm was designed to match 58 message characteristics to each user profile by following the principles of the I-Change model and maintaining the benefits of the recommender system algorithms. The mobile app resulted in a streamlined version that aimed to improve the user experience, and this system's design bridges the gap between health recommender systems and the use of behavioral change theories. This article presents a novel approach integrating recommender system technology, health behavior technology, and computer-tailored technology. Future researchers will be able to build upon the principles applied in this case study.

A Personalized Hybrid Recommendation Algorithm for Location-Based Service on Smart Campus

In order to provide location-based service for teachers and students, this paper proposes a personalized hybrid recommendation algorithm (GTRA) based on user’s geographical locations and tags. First, by analyzing the user's historical stay points and the length of stay time in different feature areas, the user's geographical adjacent user set is identified, which well solves scoring matrix sparse problem; then, we propose a tag growth strategy in which club administrators participate to give an effective solution to the problem of cold start; Finally, personalized push service was offered according to the user’s identity and the current location. Through the construction of personalized recommendation system and execution of the experiment, the algorithm proposed in this paper is superior to the traditional collaborative filtering algorithm, and it is suitable for campus application scenarios, which has some reference to the construction of smart campus in which personalized services for teachers and students are emphasized.

INVESTMENT DECISION OF TOURISM LEISURE PROJECT BASED ON COLLABORATIVE FILTERING ALGORITHM

In the era of the Big Bang, users have to spend a lot of time looking for the information they really need, and the search engine can't present information that is not described by the user. Based on the User based collaborative filtering recommendation algorithm and the collaborative filtering recommendation algorithm, this paper proposes User-CF algorithm, User-CF-1 algorithm, Item-CF algorithm, Item-CF-1 algorithm, and finally integrates four algorithms to obtain the cooperative strategy based on collaborative filtering is a hybrid recommendation algorithm, namely the Final algorithm. It has been verified that the Final algorithm can effectively solve the long tail problem in the travel strategy recommendation and greatly increase the coverage of the recommended strategy.

A hybrid recommendation algorithm–based intelligent business recommendation system

In order to solve the current issues of e–commerce recommendation systems: low accuracy, inflexible, lack personalized, etc. a solution based on hybrid recommendation algorithm is proposed, aiming at building a personalized recommendation system for e–commerce. To make up for the lack of a single recommendation algorithm, we implement the hybrid recommendation algorithm including three algorithms: content based recommendation algorithm, item based collaborative filtering recommendation algorithm and demography based recommendation algorithm. To expand recommendation dimensions, we adopt the algorithms in classification and clustering to mine the historical data of items and users. Then we do the performance evaluation and the result shows that the improved scheme has a better effect of recommendation for E–commerce.

Improving Socially-Aware Recommendation Accuracy Through Personality

In order to innovatively solve cold-start problems, research involving trust and socially aware recommender systems is currently proliferating. The relative importance of academic conferences has led to the necessity of recommender systems that seek to generate recommendations for conference attendees. In this paper, we aim to improve the recommendation accuracy of socially-aware recommender systems by proposing a linear hybrid recommender algorithm called Personality and Socially-Aware Recommender ( PerSAR ). PerSAR hybridizes the social and personality behaviours of smart conference attendees. Our recommendation methodology mainly aims to employ an algorithmic framework that computes the personality similarities and tie strengths of conference attendees so that effective and reliable recommendations can be generated for them using a relevant dataset. The experimental results substantiate that our proposed recommendation method is favorable and outperforms other related and contemporary recommendation methods and techniques.

The Design of Web Based Car Recommendation System using Hybrid Recommender Algorithm

Web based recommendations for any item is mandatory in E-commerce based web sites. This paper is about the design of web based car recommendation system using the hybrid recommender algorithm. The proposed hybrid recommender algorithm is the combination of user-to-user and item-to-item collaborative filtering method to generate the car recommendations. The user model is designed using demographic features, click data and browsing history. Item profile is built using the various attributes of car, 40 brands of car including 224 car types are used in this work. The synthetic dataset of 300 users with 10000 sessions is used to build user model. The proposed algorithm is evaluated with 100 real time users and shows the 83% accuracy in generating recommendations.

Intelligent learning system based on personalized recommendation technology

With the continuous development of networks, web-based e-learning is changing the way people acquire knowledge. An increasing number of learners are eager to acquire more knowledge through personalized and intelligent means. Based on content recommendation and collaborative filtering recommendation algorithm, this paper proposes a hybrid recommendation algorithm which can improve the efficiency of traditional recommendation algorithm. The presented research introduces the whole process of user interest model and teaching resources model, which also designs and implements the personalized network teaching resources system prototype. Finally, in comparison with the traditional recommendation algorithm, the improved hybrid recommendation algorithm has more advantages in personalized intelligent educational resources recommendation system.

Movie recommendation based on bridging movie feature and user interest

The traditional collaborative filtering algorithms have bad performance in the case of data sparsity, and are difficult to track the change of user interest. Even though many improved algorithms are proposed to solve these problems, it is still necessary for further improvement. In this paper, a novel hybrid recommendation algorithm is proposed to resolve the two issues by bridging the movie feature and user interest. In the proposed algorithm, the movie feature vector is formed based on the attributes of the movie, and is combined with the user rating matrix to generate the user interest vector. The movie feature vector and user interest vector are mutually updated in an iterative way, and then the user similarity matrix is constructed based on the user interest vector, which is usually difficult to be obtained in the case of data sparsity. Furthermore, the long-term and short-term interests are considered in the generation of the user interest vector, which aims to make the recommendation results adapt to the change of user interest. The experiments on the Movielens dataset show that the proposed algorithm outperforms some existing recommendation algorithms on recommendation accuracy.

HyRA: A Hybrid Recommendation Algorithm Focused on Smart POI. Ceutí as a Study Scenario.

Nowadays, Physical Web together with the increase in the use of mobile devices, Global Positioning System (GPS), and Social Networking Sites (SNS) have caused users to share enriched information on the Web such as their tourist experiences. Therefore, an area that has been significantly improved by using the contextual information provided by these technologies is tourism. In this way, the main goals of this work are to propose and develop an algorithm focused on the recommendation of Smart Point of Interaction (Smart POI) for a specific user according to his/her preferences and the Smart POIs’ context. Hence, a novel Hybrid Recommendation Algorithm (HyRA) is presented by incorporating an aggregation operator into the user-based Collaborative Filtering (CF) algorithm as well as including the Smart POIs’ categories and geographical information. For the experimental phase, two real-world datasets have been collected and preprocessed. In addition, one Smart POIs’ categories dataset was built. As a result, a dataset composed of 16 Smart POIs, another constituted by the explicit preferences of 200 respondents, and the last dataset integrated by 13 Smart POIs’ categories are provided. The experimental results show that the recommendations suggested by HyRA are promising.

Research of commodity recommendation workflow based on LSH algorithm

This paper joins the image-matching module into the hybrid recommendation system and constructs its workflow, which fuses image features and hybrid recommendation algorithms to improve diversity of advice result. SIFT feature extracted was used as the standard of image matching, improved LSH algorithm based on p-stable distribution to implement image matching and searching module for high dimensional and huge image set, then redesigned the workflow of existing commodity recommendation system combined with the proposed image matching module. This paper proposed an improved LSH algorithm based on p-stable distribution to finish image searching and matching. The experiment proved that the algorithm has a certain degree of optimization to improve recall rate and error rate at the same time, through the matching time and the length of hash table shows that the algorithm optimizes the memory utilization and search efficiency. The extracted SIFT feature of images is the only foundation we used when comparing different images at present. In the following research, we can try to use a variety of image features as the basis for matching to improve the reliability of the matching results.

Hybrid Recommendation Algorithm for Potential Features of Users in Social Networks Based on Swarm Intelligence

The intelligent recommendation ability of social networks is improved, a hybrid recommendation algorithm is proposed based on swarm intelligence for users' potential features in social networks. The feature extraction model of social network users is constructed, and the potential features and associated information of social network users are divided by swarm intelligence optimization technology, and the user features are learned by swarm intelligence and association rules mining. The related information of recommended items in social network is obtained, and the improvement of user item feature recommendation algorithm of social network is realized. The simulation results show that the proposed algorithm can effectively improve the accurate delivery rate of user feature recommendation in social network, and the hybrid recommendation ability for user behavior is strong, the network overhead is stable and the performance is superior

Research of Personalized Course Recommended Algorithm based on the Hybrid Recommendation

This paper presents a personalized course recommended algorithm based on the hybrid recommendation. The recommendation algorithm uses the improved NewApriori algorithm to implements the association rule recommendation, and the user-based collaborative filtering algorithm is the main part of the algorithm. The hybrid algorithm adds the weight to the recommendation result of the user-based collaborative filtering and association rule recommendation, implementing a hybrid recommendation algorithm based on both of them. It has solved the problem of data sparsity and cold-start partially and provides a academic reference for the design of high performance elective system. The experiment uses the student scores data of a college as the test set and analyzes results and recommended quality of personalized elective course. According to the results of the experimental results, the quality of the improved hybrid recommendation algorithm is better.

An Interactive Personalized Recommendation System Using the Hybrid Algorithm Model

With the rapid development of e-commerce, the contradiction between the disorder of business information and customer demand is increasingly prominent. This study aims to make e-commerce shopping more convenient, and avoid information overload, by an interactive personalized recommendation system using the hybrid algorithm model. The proposed model first uses various recommendation algorithms to get a list of original recommendation results. Combined with the customer’s feedback in an interactive manner, it then establishes the weights of corresponding recommendation algorithms. Finally, the synthetic formula of evidence theory is used to fuse the original results to obtain the final recommendation products. The recommendation performance of the proposed method is compared with that of traditional methods. The results of the experimental study through a Taobao online dress shop clearly show that the proposed method increases the efficiency of data mining in the consumer coverage, the consumer discovery accuracy and the recommendation recall. The hybrid recommendation algorithm complements the advantages of the existing recommendation algorithms in data mining. The interactive assigned-weight method meets consumer demand better and solves the problem of information overload. Meanwhile, our study offers important implications for e-commerce platform providers regarding the design of product recommendation systems.

Research on Recommendation Mode of WeChat Official Account Platform Based on Hybrid Recommendation Algorithm

Recently, the popularity of mobile socialization has promoted the development of WeChat greatly, while the WeChat official account platform is one of the most significant functions of WeChat. A hybrid recommendation algorithm combined the content-based and collaborative filter has been proposed in this paper by analyzing the hot articles posted by WeChat official accounts and the characteristics of the official account’s name. This hybrid recommendation algorithm can be used to recommend articles that have higher quality and richer content for subscriber.