Semantics and Content-Based Recommendations

Musto, Cataldo; Gemmis, Marco de; Lops, Pasquale; Narducci, Fedelucio; Semeraro, Giovanni

doi:10.1007/978-1-0716-2197-4_7

Cited by 19 publications

(7 citation statements)

References 117 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CBRS estimate an item's utility for a user based on their past preferences for similar items, determining item similarity through associated features (Ricci, Rokach, and Shapira 2022). CBRS advantages include user independence (using only the active user's ratings), transparency (explaining recommendations by listing influencing features), and the ability to recommend new, unrated items (Musto et al 2022). However, they face challenges like limited content analysis (relying on item features for recommendations), overspecialization (leading to a lack of novelty in suggestions), and the new user cold start problem (requiring sufficient user ratings for accurate recommendations) (Musto et al 2022).…”

Section: Recommendation Techniquesmentioning

confidence: 99%

Personalised Course Recommender: Linking Learning Objectives and Career Goals through Competencies

Beutling,

Spahic-Bogdanovic

2024

AAAI-SS

View full text Add to dashboard Cite

This paper presents a Knowledge-Based Recommender System (KBRS) that aims to align course recommendations with students' career goals in the field of information systems. The developed KBRS uses the European Skills, Competences, qualifications, and Occupations (ESCO) ontology, course descriptions, and a Large Language Model (LLM) such as ChatGPT 3.5 to bridge course content with the skills required for specific careers in information systems. In this context, no reference is made to the previous behavior of students. The system links course content to the skills required for different careers, adapts to students' changing interests, and provides clear reasoning for the courses proposed. An LLM is used to extract learning objectives from course descriptions and to map the promoted competency. The system evaluates the degree of relevance of courses based on the number of job-related skills supported by the learning objectives. This recommendation is supported by information that facilitates decision-making. The paper describes the system's development, methodology and evaluation and highlights its flexibility, user orientation and adaptability. It also discusses the challenges that arose during the development and evaluation of the system.

show abstract

Section: Recommendation Techniquesmentioning

confidence: 99%

Personalised Course Recommender: Linking Learning Objectives and Career Goals through Competencies

Beutling,

Spahic-Bogdanovic

2024

AAAI-SS

View full text Add to dashboard Cite

show abstract

“…Other methods used by CF are various machine learning techniques which build a summarized model of the data and predict items that the user may have an interest in [34]. CB approaches utilize user preferences over a series of item attributes to recommend additional items with similar properties [30,37]. CF and CB recommender systems need a sufficient amount of previous user ratings to generate accurate recommendations.…”

Section: User Models and Machine Learning Techniquesmentioning

confidence: 99%

“…In recent years, modeling user behavior using classical machine learning techniques has become a challenge with the large amount of information available for recommender systems and the complexity of user interactions with the system [49]. Thus, deep learning techniques have been increasingly used in recommender systems as they enable processing unstructured data and inferring hidden patterns of user preferences and item representation [30,49]. Also, deep learning techniques are able to consider descriptive information (such as text, images, audio, and video) about users and items available from various sources, to create a more reliable and accurate user model [49].…”

Section: User Models and Machine Learning Techniquesmentioning

confidence: 99%

“…It prevents CF methods from producing reliable recommendations since there is no (or too little) prior information about the user of the system which therefore cannot determine users with similar tastes to produce recommendations [44]. Moreover, since CF systems derive user preferences based on previous ratings, they must have enough ratings to create reliable user models [1,30].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Addressing Item Cold Start Problem in Collaborative Filtering-Based Recommender Systems Using Auxiliary Information

Patel

Thakkar

2022

IOT With Smart Systems

View full text Add to dashboard Cite

The initial interaction of a user with a recommender system is problematic because, in such a so-called cold start situation, the recommender system has very little information about the user, if any. Moreover, in collaborative filtering, users need to share their preferences with the service provider by rating items while in content-based filtering there is no need for such information sharing. We have recently shown that a content-based model that uses hypercube graphs can determine user preferences with a very limited number of ratings while better preserving user privacy. In this paper, we confirm these findings on the basis of experiments with more than 1,000 users in the restaurant and movie domains. We show that the proposed method outperforms standard machine learning algorithms when the number of available ratings is at most 10, which often happens, and is competitive with larger training sets. In addition, training is simple and does not require large computational efforts.

show abstract

“…As a result, the recommended items tend to be similar to items to which the user has given a high rating, which means that the users are not exposed to items that deviate from what they have already rated, even though some of those items may be of interest to them. This leads to the serendipity problem which can be defined as over-specialization in recommendations [6,18]. Users receive recommendations for similar items to those they liked in the past, whereas they may also be interested in recommendations for surprising and unexpected items [1,10,14,23].…”

Section: Introductionmentioning

confidence: 99%

Estimating serendipity in content-based recommender systems

Hertz

Kuflik

Tuval

2023

Preprint

View full text Add to dashboard Cite

Recommender systems provide personalized recommendations to their users for items and services. They do that using a model that is tailored to each user to infer their preferences based on their characteristics and previous interactions they have made with the system. Recent research suggests that users of a recommender system may like to receive suggestions that provide a pleasant surprise. In other words, a recommendation may be unexpected to the user, but it must be useful. This concept, called serendipity, is one of the aspects that have been proposed to meet user expectations for the recommendations they receive. Introducing serendipity means going beyond the `more of the same' aspect that past recommender systems are criticized for.In this article, we first show how to estimate user preferences based on ratings they have done in the past in a content-based recommender system.This estimation allows us to measure the relevance of a recommendation. We then determine the item attributes that play an important role in the relevance measure.Experiments in the movie domain show that the greater the relevance of a recommendation, the more the users seem willing to discover items having attributes with which they are not familiar, as long as these do not play an important role in their ratings.

show abstract

Semantics and Content-Based Recommendations

Cited by 19 publications

References 117 publications

Personalised Course Recommender: Linking Learning Objectives and Career Goals through Competencies

Personalised Course Recommender: Linking Learning Objectives and Career Goals through Competencies

Addressing Item Cold Start Problem in Collaborative Filtering-Based Recommender Systems Using Auxiliary Information

Estimating serendipity in content-based recommender systems

Contact Info

Product

Resources

About