Content-Based approaches for Cold-Start Job Recommendations

Bianchi, Mattia; Cesaro, Federico; Ciceri, Filippo; Dagrada, Mattia; Gasparin, Alberto; Grattarola, Daniele; Inajjar, Ilyas; Metelli, Alberto Maria; Cella, Leonardo

doi:10.1145/3124791.3124793

Cited by 13 publications

(12 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Answering question Q4 defined in Table 1, if we focus on the different filtering methods used by the analyzed systems, we can see a tendency in the use of the collaborative filter. This is not a common result; normally content-based filters are used as starting filtering methods [32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47], and once a considerable amount of data from the users are obtained, collaborative filters are launched. This might be because some studies were started with some initial data about the users.…”

Section: Discussionmentioning

confidence: 99%

Systematic Review of Nutritional Recommendation Systems

2021

View full text Add to dashboard Cite

In recent years, the promotion of healthy habits, and especially diet-oriented habits, has been one of the priority interests of our society. There are many apps created to count calories based on what we eat, or to estimate calorie consumption according to the sport we do, or to recommend recipes, but very few are capable of giving personalized recommendations. This review tries to see what studies exist and what recommendation systems are used for this purpose, over the last 5 years in the main databases. Among the results obtained, it is observed that the existing works focus on the recommendation system (usually collaborative filtering), and not so much on the description of the data or the sample analyzed; the indices used for the calculation of calories or nutrients are not specified. Therefore, it is necessary to work with open data, or well-described data, which allows the experience to be reproduced by third parties, or at least to be comparable. In recent years, the promotion of healthy habits, and especially diet-oriented habits, has been one of the priority interests of our society.

show abstract

Section: Discussionmentioning

confidence: 99%

Systematic Review of Nutritional Recommendation Systems

2021

View full text Add to dashboard Cite

show abstract

“…However, most of them are not explicitly tailored for solving user cold-start recommendation, e.g., new registered users only have very few interactions [36]. Previous methods [3,12] mainly try to incorporate side-information into CF for alleviating this problem.…”

Section: User Cold-start Recommendationmentioning

confidence: 99%

“…To address user cold-start recommendation, early methods [3,28] focus on using side-information, i.e., user profiles and item contents, to infer the preferences of cold-start users. Additionally, many hybrid models integrating side-information into collaborative filtering (CF) are also proposed [24,47].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Task-adaptive Neural Process for User Cold-Start Recommendation

Lin

Zhou

et al. 2021

Proceedings of the Web Conference 2021

View full text Add to dashboard Cite

User cold-start recommendation is a long-standing challenge for recommender systems due to the fact that only a few interactions of cold-start users can be exploited. Recent studies seek to address this challenge from the perspective of meta learning, and most of them follow a manner of parameter initialization, where the model parameters can be learned by a few steps of gradient updates. While these gradient-based meta-learning models achieve promising performances to some extent, a fundamental problem of them is how to adapt the global knowledge learned from previous tasks for the recommendations of cold-start users more effectively.In this paper, we develop a novel meta-learning recommender called task-adaptive neural process (TaNP). TaNP is a new member of the neural process family, where making recommendations for each user is associated with a corresponding stochastic process. TaNP directly maps the observed interactions of each user to a predictive distribution, sidestepping some training issues in gradient-based meta-learning models. More importantly, to balance the trade-off between model capacity and adaptation reliability, we introduce a novel task-adaptive mechanism. It enables our model to learn the relevance of different tasks and customize the global knowledge to the task-related decoder parameters for estimating user preferences. We validate TaNP on multiple benchmark datasets in different experimental settings. Empirical results demonstrate that TaNP yields consistent improvements over several state-ofthe-art meta-learning recommenders. CCS CONCEPTS• Information systems → Recommender systems; • Computing methodologies → Neural networks.

show abstract

“…Moreover, we discard impression interactions as they are created when XING shows a job to a user. As stated by Bianchi et al (2017), an impression does not imply that the user has interacted with the job. The dataset consists of interactions from a period of three months (from November 6, 2016, until February 3, 2017.…”

Section: Recsys17mentioning

confidence: 99%

Using autoencoders for session-based job recommendations

Lacic

Reiter-Haas²,

Kowald

et al. 2020

User Model User-Adap Inter

View full text Add to dashboard Cite

In this work, we address the problem of providing job recommendations in an online session setting, in which we do not have full user histories. We propose a recommendation approach, which uses different autoencoder architectures to encode sessions from the job domain. The inferred latent session representations are then used in a k-nearest neighbor manner to recommend jobs within a session. We evaluate our approach on three datasets, (1) a proprietary dataset we gathered from the Austrian student job portal Studo Jobs, (2) a dataset released by XING after the RecSys 2017 Challenge and (3) anonymized job applications released by CareerBuilder in 2012. Our results show that autoencoders provide relevant job recommendations as well as maintain a high coverage and, at the same time, can outperform state-of-the-art session-based recommendation techniques in terms of system-based and sessionbased novelty.

show abstract

Content-Based approaches for Cold-Start Job Recommendations

Cited by 13 publications

References 11 publications

Systematic Review of Nutritional Recommendation Systems

Systematic Review of Nutritional Recommendation Systems

Task-adaptive Neural Process for User Cold-Start Recommendation

Using autoencoders for session-based job recommendations

Contact Info

Product

Resources

About