The  Definition of  Novelty in  Recommendation  System

Zhang, Liang

doi:10.25103/jestr.063.25

Cited by 29 publications

(18 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Taking into consideration the level of novelty, serendipity, and dynamism of this algorithm, we can tell that the algorithm was really novel and serendipitous given that the rate was 84% at average. e level of novelty was better as compared to other novel algorithms which are found in the literature such as [40,41] which experience a maximum of 77%. e algorithm's recommendations were also changing with time sessions especially the four time sessions (morning, afternoon, evening, and midnight), which means it was considering time context very well.…”

Section: Noveltymentioning

confidence: 80%

“…Novelty determines how unknown recommended items are to a user. e novelty of a retrieval set has been defined with respect to the end user as the proportion of known and unknown relevant items in the recommended list [40].…”

Section: Noveltymentioning

confidence: 99%

“…e best way to measure novelty and serendipity is asking users whether they were already familiar to a recommended item. However, this is impracticable in offline experiments [40]. It is for this reason that we took an online experimental approach to dealing with real users.…”

Section: Noveltymentioning

confidence: 99%

See 2 more Smart Citations

Holistic User Context-Aware Recommender Algorithm

Kavu

Dube

Raeth

2019

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Existing recommender algorithms lack dynamism, human focus, and serendipitous recommendations. The literature indicates that the context of a user influences user decisions, and when incorporated in recommender systems (RSs), novel and serendipitous recommendations can be realized. This article shows that social, cultural, psychological, and economic contexts of a user influence user traits or decisions. The article demonstrates a novel approach of incorporating holistic user context-aware knowledge in an algorithm to solve the highlighted problems. Web content mining and collaborative filtering approaches were used to develop a holistic user context-aware (HUC) algorithm. The algorithm was evaluated on a social network using online experimental evaluations. The algorithm demonstrated dynamism, novelty, and serendipity with an average of 84% novelty and 85% serendipity.

show abstract

Section: Noveltymentioning

confidence: 80%

Section: Noveltymentioning

confidence: 99%

See 1 more Smart Citation

Holistic User Context-Aware Recommender Algorithm

Kavu

Dube

Raeth

2019

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…Novelty is one of the important indicators of user satisfaction as it helps users in the decision-making process [59]. OPCR provided the users with recommendations that included novel items which were not expected because ontology mapping is able to link all of the attributes in the course profiles and user profiles.…”

Section: ) Noveltymentioning

confidence: 99%

Ontology-Based Personalized Course Recommendation Framework

et al. 2019

View full text Add to dashboard Cite

Choosing a higher education course at university is not an easy task for students. A wide range of courses are offered by the individual universities whose delivery mode and entry requirements differ. A personalized recommendation system can be an effective way of suggesting the relevant courses to the prospective students. This paper introduces a novel approach that personalizes course recommendations that will match the individual needs of users. The proposed approach developed a framework of an ontologybased hybrid-filtering system called the ontology-based personalized course recommendation (OPCR). This approach aims to integrate the information from multiple sources based on the hierarchical ontology similarity with a view to enhancing the efficiency and the user satisfaction and to provide students with appropriate recommendations. The OPCR combines collaborative-based filtering with content-based filtering. It also considers familiar related concepts that are evident in the profiles of both the student and the course, determining the similarity between them. Furthermore, OPCR uses an ontology mapping technique, recommending jobs that will be available following the completion of each course. This method can enable students to gain a comprehensive knowledge of courses based on their relevance, using dynamic ontology mapping to link the course profiles and student profiles with job profiles. Results show that a filtering algorithm that uses hierarchically related concepts produces better outcomes compared to a filtering method that considers only keyword similarity. In addition, the quality of the recommendations is improved when the ontology similarity between the items' and the users' profiles were utilized. This approach, using a dynamic ontology mapping, is flexible and can be adapted to different domains. The proposed framework can be used to filter the items for both postgraduate courses and items from other domains.

show abstract

“…could lead to a decline in recommendation accuracy as users lose interest over time. Therefore, we compared the diversity [4] and novelty [31] of recommended items. The proposed model was able to recommend videos with better novelty and diversity, and was the closest to NUBPR-O model (only a 2.4% and 3.8% novelty and diversity drop).…”

Section: Diversity and Noveltymentioning

confidence: 99%

CnGAN: Generative Adversarial Networks for Cross-network user preference generation for non-overlapped users

Perera

Zimmermann

2019

The World Wide Web Conference

View full text Add to dashboard Cite

A major drawback of cross-network recommender solutions is that they can only be applied to users that are overlapped across networks. Thus, the non-overlapped users, which form the majority of users are ignored. As a solution, we propose CnGAN, a novel multi-task learning based, encoder-GAN-recommender architecture. The proposed model synthetically generates source network user preferences for non-overlapped users by learning the mapping from target to source network preference manifolds. The resultant user preferences are used in a Siamese network based neural recommender architecture. Furthermore, we propose a novel userbased pairwise loss function for recommendations using implicit interactions to better guide the generation process in the multi-task learning environment. We illustrate our solution by generating user preferences on the Twitter source network for recommendations on the YouTube target network. Extensive experiments show that the generated preferences can be used to improve recommendations for non-overlapped users. The resultant recommendations achieve superior performance compared to the state-of-the-art cross-network recommender solutions in terms of accuracy, novelty and diversity.

show abstract

The Definition of Novelty in Recommendation System

Cited by 29 publications

References 25 publications

Holistic User Context-Aware Recommender Algorithm

Holistic User Context-Aware Recommender Algorithm

Ontology-Based Personalized Course Recommendation Framework

CnGAN: Generative Adversarial Networks for Cross-network user preference generation for non-overlapped users

Contact Info

Product

Resources

About