An extensive study on the evolution of context-aware personalized travel recommender systems
یک مطالعه گسترده در مورد تکامل سیستمهای توصیه گر سفر شخصی آگاه از متن-2020
Ever since the beginning of civilization, travel for various causes exists as an essential part of human life so as travel recommendations, though the early form of recommendations were the accrued experiences shared by the community. Modern recommender systems evolved along with the growth of Information Technology and are contributing to all industry and service segments inclusive of travel and tourism. The journey started with generic recommender engines which gave way to personalized recommender systems and further advanced to contextualized personalization with advent of artificial intelligence. Current era is also witnessing a boom in social media usage and the social media big data is acting as a critical input for various analytics with no exception for recommender systems. This paper details about the study conducted on the evolution of travel recommender systems, their features and current set of limitations. We also discuss on the key algorithms being used for classification and recommendation processes and metrics that can be used to evaluate the performance of the algorithms and thereby the recommenders.
Keywords: Recommender system | Personalization | Context aware | Big data | Travel and tourism
Towards a more reliable privacy-preserving recommender system
به سمت یک سیستم توصیه گر برای حفظ حریم خصوصی قابل اطمینان تر-2019
This paper proposes a privacy-preserving distributed recommendation framework, Secure Distributed Collaborative Filtering (SDCF), to preserve the privacy of value, model and ex- istence altogether. That says, not only the ratings from the users to the items, but also the existence of the ratings as well as the learned recommendation model are kept pri- vate in our framework. Our solution relies on a distributed client-server architecture and a two-stage Randomized Response algorithm, along with an implementation on the pop- ular recommendation model, Matrix Factorization (MF). We further prove SDCF to meet the guarantee of Differential Privacy so that clients are allowed to specify arbitrary privacy levels. Experiments conducted on numerical rating prediction and one-class rating action prediction exhibit that SDCF does not sacriﬁce too much accuracy for privacy.© 2019 Elsevier Inc. All rights reserved.
Keywords: Privacy-preserving recommendation | Differential privacy | Secure distributed matrix factorization | Randomized response algorithms
A sub-one quasi-norm-based similarity measure for collaborative filtering in recommender systems
اندازه گیری شباهت sub-one مبتنی بر شبه هنجار برای فیلتر مشترک در سیستم های توصیه گر-2019
Collaborative filtering (CF) is one of the most successful approaches for an online store to make personalized recommendations through its recommender systems. A neighborhoodbased CF method makes recommendations to a target customer based on the similar preference of the target customer and those in the database. Similarity measuring between users directly contributes to an effective recommendation. In this paper, we propose a sub-one quasi-norm-based similarity measure for collaborative filtering in a recommender system. The proposed similarity measure shows its advantages over those commonly used similarity measures in the literature by making better use of rating values and deemphasizing the dissimilarity between users. Computational experiments using various real-life datasets clearly indicate the superiority of the proposed similarity measure, no matter in fully co-rated, sparsely co-rated or cold-start scenarios.
Keywords: Recommender system | Collaborative filtering | Neighborhood-based CF | Similarity measure | p quasi-norm
A meta-learning recommender system for hyperparameter tuning: Predicting when tuning improves SVM classifiers
یک سیستم توصیه گر متا یادگیری برای تنظیم hyperparameter : پیش بینی زمانیکه تنظیم باعث بهبود طبقه بندی های SVM می شود-2019
For many machine learning algorithms, predictive performance is critically affected by the hyperparameter values used to train them. However, tuning these hyperparameters can come at a high computational cost, especially on larger datasets, while the tuned settings do not always significantly outperform the default values. This paper proposes a recom- mender system based on meta-learning to identify exactly when it is better to use de- fault values and when to tune hyperparameters for each new dataset. Besides, an in-depth analysis is performed to understand what they take into account for their decisions, pro- viding useful insights. An extensive analysis of different categories of meta-features, meta- learners, and setups across 156 datasets is performed. Results show that it is possible to accurately predict when tuning will significantly improve the performance of the induced models. The proposed system reduces the time spent on optimization processes, without reducing the predictive performance of the induced models (when compared with the ones obtained using tuned hyperparameters). We also explain the decision-making pro- cess of the meta-learners in terms of linear separability-based hypotheses. Although this analysis is focused on the tuning of Support Vector Machines, it can also be applied to other algorithms, as shown in experiments performed with decision trees.
Keywords: Meta-learning | Recommender system | Tuning recommendation | Hyperparameter tuning | Support vector machines
Slanderous user detection with modified recurrent neural networks in recommender system
تشخیص کاربر مخدوش با شبکه های عصبی بازرخدادگر اصلاح شده در سیستم توصیه گر-2019
We focus on how to tackle a unique multi-view unsupervised issue: slanderous user de- tection, with recurrent neural networks to benefit recommender systems. In real-world recommender systems, some consumers always give fake reviews and low ratings to the items they bought on purpose. In order to ensure their profits, these slanderous users make a semantic gap between their ratings and reviews to avoid detection, which makes slanderous user detection a more difficult problem. On some occasions, they give a false low rating with a positive review which confuse recommender systems, and vice versa. To address the above problem, in this paper, we propose a novel recommendation frame- work: Slanderous user Detection Recommender System (SDRS). In SDRS, we design a Hier- archical Dual-Attention recurrent Neural network (HDAN) with a modified GRU (mGRU) to compute an opinion level for reviews. Then a joint filtering method is proposed to catch the gap between ratings and reviews. With joint filtering, slanderous users can be de- tected and omitted. Finally, a modified non-negative matrix factorization is proposed to make recommendations. Extensive experiments are conducted in four datasets: Amazon, Yelp, Taobao, and Jingdong, in which the results demonstrate that our proposed method can detect slanderous users and make accurate recommendations in a uniform framework. Also, with slanderous user detection, some state-of-the-art recommendation systems can be benefited.
Keywords: Slanderous user detection | Recommender systems | Recurrent neural networks
Gaussian-Gamma collaborative filtering: A hierarchical Bayesian model for recommender systems
فیلتر مشارکتی گاوسین-گاما: یک مدل سلسله مراتبی بیزی برای سیستمهای توصیه گر-2019
The traditional collaborative filtering (CF) suffers from two key challenges, namely, the normal assumption that it is not robust, and it is difficult to set in advance the penalty terms of the latent features. We therefore propose a hierarchical Bayesian model-based CF and the related inference algorithm. Specifically, we impose a Gaussian-Gamma prior on the ratings, and the latent features. We show the model is more robust, and the penalty terms can be adapted automatically in the inference. We use Gibbs sampler for the inference and provide a statistical explanation. We verify the performance using both synthetic and real datasets.
Keywords: Gaussian-Gamma distribution | Recommender system | Hierarchical Bayesian model | Gibbs Sampling | Performance evaluation
Towards autonomy: A recommender system for the determination of trim and flight parameters for Seagliders
به سمت استقلال: سیستم توصیه گر برای تعیین پارامترهای تریم و پرواز برای Seagliders-2019
Currently, pilots maximise the performance of Seaglider underwater gliders by manually selecting their set-up parameters. Building on existing procedures based on the assumption of steady-state motions, a recommender system for the trim and flight parameters has been developed to aid trainee pilots and enable round-the-clock operations. The system has been validated with data from 12 missions run in waters off the United Kingdom and Australia, representative of a range of oceanographic conditions. The recommended trim parameters present a maximum difference of 14% from the values selected by the pilots, whereas pilots are found not to change the flight parameters. Additionally, suggestions are made to improve operational practices to further improve the accuracy of the recommender system. As a result, the developed system is expected to greatly help trainee pilots achieve expertise in a much smaller time frame than standard practice. Additionally, thanks to its high precision, the recommender system can be used to autonomously select the trim and flight parameters of Seagliders for night operations in the future.
Keywords: Autonomous underwater vehicle (AUV) | Underwater glider | System identification | Recommender system
A personalized electricity tariff recommender system based on advanced metering infrastructure and collaborative filtering
یک سیستم توصیه گر تعرفه برق شخصی مبتنی بر زیرساخت های اندازه گیری پیشرفته و فیلترهای مشترک-2019
Deregulation of electricity retail markets and the advancement of energy informatics have been supporting the transition of electricity retail business into an electronic business, where different electricity retailers can provide different electricity tariff plans to end users through digital media. In this context, end users are facing an information filtering challenge of choosing the most suitable tariff plans from a set of candidate tariff plans. This paper proposes a new personalized recommendation system that makes intelligent electricity tariff recommendations to end users. The proposed approach starts by collecting a group of end users’ electricity consumption profiles through the advanced metering infrastructure and, based on this information, it infers the preference of individual users on each tariff plan. Based on the inferred preference degree, a new matrix factorization is established based on a collaborative filtering algorithm that is capable of recommending most suitable tariff plans to an arbitrary target user. The proposed recommendation system is validated against a number of scenarios that are generated based on simulated tariff plan sets and on a modified Australian “Smart Grid, Smart City” dataset.
Keywords: Collaborative filtering | Power market | Recommendation system | Demand side management | Smart grid
GimmeHop: A recommender system for mobile devices using ontology reasoners and fuzzy logic
GimmeHop: یک سیستم توصیه گر برای دستگاه های همراه سیار با استفاده از استدلال هستی شناسی و منطق فازی-2019
This paper describes GimmeHop, a beer recommender system for Android mobile devices using fuzzy ontologies to represent the relevant knowledge and semantic reasoners to infer implicit knowledge. GimmeHop use fuzzy quantifiers to deal with incomplete data, fuzzy hedges to deal with the user context, and aggregation operators to manage user preferences. The results of our evaluation measure empirically the data traffic and the running time in the case of remote reasoning, the size of the ontologies that can be locally dealt with in a mobile device in the case of local reasoning, and the quality of the automatically computed linguistic values supported in the user queries
Keywords:Fuzzy ontologies | Aggregation | Fuzzy quantifiers | Recommender systems
Deep latent factor model for collaborative filtering
مدل فاکتور نهفته عمیق برای فیلتر مشارکتی-2019
Latent factor models have been used widely in collaborative filtering based recommender systems. In re- cent years, deep learning has been successful in solving a wide variety of machine learning problems. Mo- tivated by the success of deep learning, we propose a deeper version of latent factor model. Experiments on benchmark datasets shows that our proposed technique significantly outperforms all state-of-the-art collaborative filtering techniques.
Keywords: Deep learning | Latent semantic analysis | Collaborative filtering | Recommender systems