STARec: Adaptive Learning with Spatiotemporal and Activity Influence for POI Recommendation

2022 ◽  
Vol 40 (4) ◽  
pp. 1-40
Author(s):  
Weiyu Ji ◽  
Xiangwu Meng ◽  
Yujie Zhang
Keyword(s):  
Adaptive Learning ◽  
Social Preferences ◽  
Close Correlation ◽  
User Preferences ◽  
Superior Performance ◽  
Individual User ◽  
User Interests ◽  
Poi Recommendation ◽  
The Social ◽  
Real World Datasets

POI recommendation has become an essential means to help people discover attractive places. Intuitively, activities have an important impact on users’ decision-making, because users select POIs to attend corresponding activities. However, many existing studies ignore the social motivation of user behaviors and regard all check-ins as influenced only by individual user interests. As a result, they cannot model user preferences accurately, which degrades recommendation effectiveness. In this article, from the perspective of activities, this study proposes a probabilistic generative model called STARec. Specifically, based on the social effect of activities, STARec defines users’ social preferences as distinct from their individual interests and combines these with individual user activity interests to effectively depict user preferences. Moreover, the inconsistency between users’ social preferences and their decisions is modeled. An activity frequency feature is introduced to acquire accurate user social preferences because of close correlation between these and the key impact factor of corresponding check-ins. An alias sampling-based training method was used to accelerate training. Extensive experiments were conducted on two real-world datasets. Experimental results demonstrated that the proposed STARec model achieves superior performance in terms of high recommendation accuracy, robustness to data sparsity, effectiveness in handling cold-start problems, efficiency, and interpretability.

scholarly journals MFNP: A Meta-optimized Model for Few-shot Next POI Recommendation

2021 ◽  
Author(s):  
Huimin Sun ◽  
Jiajie Xu ◽  
Kai Zheng ◽  
Pengpeng Zhao ◽  
Pingfu Chao ◽  
...  
Keyword(s):  
State Of The Art ◽  
Cold Start ◽  
User Preferences ◽  
Location Based Services ◽  
Point Of Interest ◽  
Poi Recommendation ◽  
Meta Learning ◽  
Optimized Model ◽  
Real World Datasets ◽  
Cold Start Problem

Next Point-of-Interest (POI) recommendation is of great value for location-based services. Existing solutions mainly rely on extensive observed data and are brittle to users with few interactions. Unfortunately, the problem of few-shot next POI recommendation has not been well studied yet. In this paper, we propose a novel meta-optimized model MFNP, which can rapidly adapt to users with few check-in records. Towards the cold-start problem, it seamlessly integrates carefully designed user-specific and region-specific tasks in meta-learning, such that region-aware user preferences can be captured via a rational fusion of region-independent personal preferences and region-dependent crowd preferences. In modelling region-dependent crowd preferences, a cluster-based adaptive network is adopted to capture shared preferences from similar users for knowledge transfer. Experimental results on two real-world datasets show that our model outperforms the state-of-the-art methods on next POI recommendation for cold-start users.

scholarly journals Where to Go Next: Modeling Long- and Short-Term User Preferences for Point-of-Interest Recommendation

2020 ◽  
Vol 34 (01) ◽  
pp. 214-221 ◽  
Author(s):  
Ke Sun ◽  
Tieyun Qian ◽  
Tong Chen ◽  
Yile Liang ◽  
Quoc Viet Hung Nguyen ◽  
...  
Keyword(s):  
State Of The Art ◽  
User Preferences ◽  
Short Term ◽  
Preference Modeling ◽  
Point Of Interest ◽  
Proposed Model ◽  
Poi Recommendation ◽  
Novel Method ◽  
Real World Datasets

Point-of-Interest (POI) recommendation has been a trending research topic as it generates personalized suggestions on facilities for users from a large number of candidate venues. Since users' check-in records can be viewed as a long sequence, methods based on recurrent neural networks (RNNs) have recently shown promising applicability for this task. However, existing RNN-based methods either neglect users' long-term preferences or overlook the geographical relations among recently visited POIs when modeling users' short-term preferences, thus making the recommendation results unreliable. To address the above limitations, we propose a novel method named Long- and Short-Term Preference Modeling (LSTPM) for next-POI recommendation. In particular, the proposed model consists of a nonlocal network for long-term preference modeling and a geo-dilated RNN for short-term preference learning. Extensive experiments on two real-world datasets demonstrate that our model yields significant improvements over the state-of-the-art methods.

Social experience influences preferences in black ducks and mallards

1989 ◽  
Vol 67 (6) ◽  
pp. 1434-1438 ◽  
Author(s):  
Lynn M. Brodsky ◽  
C. Davison Ankney ◽  
Darrell G. Dennis
Keyword(s):  
Social Preferences ◽  
Anas Platyrhynchos ◽  
Social Experience ◽  
The Other ◽  
Captive Population ◽  
Male And Female ◽  
Potential Mate ◽  
The Social ◽  
Opposite Sex ◽  
Males And Females

The influence of social experience on the preferences for a potential mate in a captive population of black ducks, Anas rubripes, and mallards, Anas platyrhynchos, was examined. Birds were reared from hatching with conspecifics (i.e., female black ducks with male black ducks, female mallards with male mallards), or were cross-fostered with the other species (i.e., female black ducks with male mallards, female mallards with male black ducks). Preferences of individuals were tested in a chamber containing caged black ducks and mallards of the opposite sex. In over 90% (100/109) of the trials, males and females preferred the species that they were raised with since hatching, whether they were of the same species or not. These results demonstrate that social experience influences the social preferences of male and female black ducks and mallards.

Political Competition, Partisanship and Interpersonal Trust in Electoral Democracies

2016 ◽  
Vol 48 (1) ◽  
pp. 115-139 ◽  
Author(s):  
Ryan E. Carlin ◽  
Gregory J. Love
Keyword(s):  
Party System ◽  
Social Preferences ◽  
Political Competition ◽  
Interpersonal Trust ◽  
The United States ◽  
Behavioral Experiments ◽  
Minimal Group ◽  
System A ◽  
The Social ◽  
Osama Bin Laden

How does democratic politics inform the interdisciplinary debate on the evolution of human co-operation and the social preferences (for example, trust, altruism and reciprocity) that support it? This article advances a theory of partisan trust discrimination in electoral democracies based on social identity, cognitive heuristics and interparty competition. Evidence from behavioral experiments in eight democracies show ‘trust gaps’ between co- and rival partisans are ubiquitous, and larger than trust gaps based on the social identities that undergird the party system. A natural experiment found that partisan trust gaps in the United States disappeared immediately following the killing of Osama bin Laden. But observational data indicate that partisan trust gaps track with perceptions of party polarization in all eight cases. Finally, the effects of partisanship on trust outstrip minimal group treatments, yet minimal-group effects are on par with the effects of most treatments for ascriptive characteristics in the literature. In sum, these findings suggest political competition dramatically shapes the salience of partisanship in interpersonal trust, the foundation of co-operation.

scholarly journals Unsupervised Anomaly Detection with Distillated Teacher-Student Network Ensemble

Entropy ◽  
2021 ◽  
Vol 23 (2) ◽  
pp. 201
Author(s):  
Qinfeng Xiao ◽  
Jing Wang ◽  
Youfang Lin ◽  
Wenbo Gongsa ◽  
Ganghui Hu ◽  
...  
Keyword(s):  
Anomaly Detection ◽  
Multivariate Data ◽  
Failure Detection ◽  
Superior Performance ◽  
Detection Algorithms ◽  
Teacher Student ◽  
Model Complex ◽  
Unsupervised Anomaly Detection ◽  
Real World Datasets ◽  
Complex Features

We address the problem of unsupervised anomaly detection for multivariate data. Traditional machine learning based anomaly detection algorithms rely on specific assumptions of normal patterns and fail to model complex feature interactions and relations. Recently, existing deep learning based methods are promising for extracting representations from complex features. These methods train an auxiliary task, e.g., reconstruction and prediction, on normal samples. They further assume that anomalies fail to perform well on the auxiliary task since they are never trained during the model optimization. However, the assumption does not always hold in practice. Deep models may also perform the auxiliary task well on anomalous samples, leading to the failure detection of anomalies. To effectively detect anomalies for multivariate data, this paper introduces a teacher-student distillation based framework Distillated Teacher-Student Network Ensemble (DTSNE). The paradigm of the teacher-student distillation is able to deal with high-dimensional complex features. In addition, an ensemble of student networks provides a better capability to avoid generalizing the auxiliary task performance on anomalous samples. To validate the effectiveness of our model, we conduct extensive experiments on real-world datasets. Experimental results show superior performance of DTSNE over competing methods. Analysis and discussion towards the behavior of our model are also provided in the experiment section.

scholarly journals SetRank: A Setwise Bayesian Approach for Collaborative Ranking from Implicit Feedback

2020 ◽  
Vol 34 (04) ◽  
pp. 6127-6136
Author(s):  
Chao Wang ◽  
Hengshu Zhu ◽  
Chen Zhu ◽  
Chuan Qin ◽  
Hui Xiong
Keyword(s):  
Bayesian Approach ◽  
Posterior Probability ◽  
State Of The Art ◽  
User Preferences ◽  
Implicit Feedback ◽  
Pairwise Preference ◽  
Entire List ◽  
Collaborative Ranking ◽  
Real World Datasets ◽  
Made In

The recent development of online recommender systems has a focus on collaborative ranking from implicit feedback, such as user clicks and purchases. Different from explicit ratings, which reflect graded user preferences, the implicit feedback only generates positive and unobserved labels. While considerable efforts have been made in this direction, the well-known pairwise and listwise approaches have still been limited by various challenges. Specifically, for the pairwise approaches, the assumption of independent pairwise preference is not always held in practice. Also, the listwise approaches cannot efficiently accommodate “ties” due to the precondition of the entire list permutation. To this end, in this paper, we propose a novel setwise Bayesian approach for collaborative ranking, namely SetRank, to inherently accommodate the characteristics of implicit feedback in recommender system. Specifically, SetRank aims at maximizing the posterior probability of novel setwise preference comparisons and can be implemented with matrix factorization and neural networks. Meanwhile, we also present the theoretical analysis of SetRank to show that the bound of excess risk can be proportional to √M/N, where M and N are the numbers of items and users, respectively. Finally, extensive experiments on four real-world datasets clearly validate the superiority of SetRank compared with various state-of-the-art baselines.

scholarly journals Evolutionary Social Poisson Factorizationfor Temporal Recommendation

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
ChunYan Yin ◽  
YongHeng Chen ◽  
Wanli Zuo
Keyword(s):  
Recommendation Systems ◽  
Predictive Performance ◽  
User Preferences ◽  
Social Model ◽  
User Preference ◽  
Latent Factors ◽  
Factorization Model ◽  
Real World Datasets ◽  
Item Attributes ◽  
The Impact

AbstractPreference-based recommendation systems analyze user-item interactions to reveal latent factors that explain our latent preferences for items and form personalized recommendations based on the behavior of others with similar tastes. Most of the works in the recommendation systems literature have been developed under the assumption that user preference is a static pattern, although user preferences and item attributes may be changed through time. To achieve this goal, we develop an Evolutionary Social Poisson Factorization (EPF$$\_$$ _ Social) model, a new Bayesian factorization model that can effectively model the smoothly drifting latent factors using Conjugate Gamma–Markov chains. Otherwise, EPF$$\_$$ _ Social can obtain the impact of friends on social network for user’ latent preferences. We studied our models with two large real-world datasets, and demonstrated that our model gives better predictive performance than state-of-the-art static factorization models.

scholarly journals Bought and Sold: Exploring the Effects of Big Data on User Agency and Commodification

2021 ◽  
Author(s):  
Kristia M. Pavlakos
Keyword(s):  
Social Sciences ◽  
Big Data ◽  
Data Privacy ◽  
Large Data ◽  
Data Sets ◽  
Privacy Regulation ◽  
Scholarly Literature ◽  
User Interests ◽  
The Social ◽  
The Relationship

Big Data1is a phenomenon that has been increasingly studied in the academy in recent years, especially in technological and scientific contexts. However, it is still a relatively new field of academic study; because it has been previously considered in mainly technological contexts, more attention needs to be drawn to the contributions made in Big Data scholarship in the social sciences by scholars like Omar Tene and Jules Polonetsky, Bart Custers, Kate Crawford, Nick Couldry, and Jose van Dijk. The purpose of this Major Research Paper is to gain insight into the issues surrounding privacy and user rights, roles, and commodification in relation to Big Data in a social sciences context. The term “Big Data” describes the collection, aggregation, and analysis of large data sets. While corporations are usually responsible for the analysis and dissemination of the data, most of this data is user generated, and there must be considerations regarding the user’s rights and roles. In this paper, I raise three main issues that shape the discussion: how users can be more active agents in data ownership, how consent measures can be made to actively reflect user interests instead of focusing on benefitting corporations, and how user agency can be preserved. Through an analysis of social sciences scholarly literature on Big Data, privacy, and user commodification, I wish to determine how these concepts are being discussed, where there have been advancements in privacy regulation and the prevention of user commodification, and where there is a need to improve these measures. In doing this, I hope to discover a way to better facilitate the relationship between data collectors and analysts, and user-generators. 1 While there is no definitive resolution as to whether or not to capitalize the term “Big Data”, in capitalizing it I chose to conform with such authors as boyd and Crawford (2012), Couldry and Turow (2014), and Dalton and Thatcher (2015), who do so in the scholarly literature.

A New Look into the Determinants of the Ecological Discount Rate: Disentangling Social Preferences

2012 ◽  
Vol 12 (1) ◽  
Author(s):  
Luciana Echazu ◽  
Diego Nocetti ◽  
William T. Smith
Keyword(s):  
Term Structure ◽  
Discount Rate ◽  
Environmental Quality ◽  
Social Preferences ◽  
Inequality Aversion ◽  
The Social ◽  
Representative Consumer ◽  
Change Mitigation ◽  
Over Time

Abstract How should changes in environmental quality occurring in the future be discounted? To answer this question we consider a model of “ecological discounting”, where the representative consumer has a utility function defined over two attributes, consumption and environmental quality, which evolve stochastically over time. We characterize the determinants of the social discount rate and its behavior over time using a preference structure that disentangles attitudes towards intertemporal inequality, attitudes towards risk, and tastes over consumption and environmental quality. We show that the degree of substitutability between consumption and environmental quality, the degree of risk aversion, the degree of inequality aversion, and the rate at which these attitudes change as natural and man-made resources evolve over time are all important aspects of the ecological discount rate and its term structure. Our analysis suggests that over medium and long term horizons the ecological discount rate should be below the rate of time preference, supporting recent proposals for immediate action towards climate change mitigation.

The Future of ICT/LKT in the Teaching and Learning of Social Sciences, Geography, and History

2020 ◽  
pp. 268-291
Keyword(s):  
Social Sciences ◽  
Artificial Intelligence ◽  
Adaptive Learning ◽  
Teaching And Learning ◽  
Industrial Revolution ◽  
Learning Goals ◽  
Fourth Industrial Revolution ◽  
Online Tools ◽  
The Social ◽  
New Challenges

Most experts consider that society has entered in a Fourth Industrial Revolution that implies ubiquitous changes characterized by a fusion of technologies that is blurring the lines that differentiate physical, digital, and biological spheres. This implies to open a door to important changes in the teaching and learning of the social sciences, geography, and history. Regarding this, it is necessary that both citizens and organizations develop new skills. Artificial intelligence as education technology is possible due to digital and online tools. Adaptive learning, meanwhile, is related to artificial intelligence, personalizing the learning and offering contents adapted to students. New challenges in the teaching of social sciences extends beyond the learning of facts and events. As a result of changes in society of Fourth Industrial Revolution, thinking-based learning (TBL) with the support of learning and knowledge technologies (LKT), creativity, critical thinking, and cooperation are some of the essential learning goals to participate in society.

Sign in / Sign up

Export Citation Format

Share Document