Task In Recommender Systems Research Articles

MARAN: Supporting awareness of users’ routines and preferences for next POI recommendation based on spatial aggregation

Next point-of-interest (POI) recommendation has emerged as an essential task in recommender systems with the rapid development of location-based social networks (LBSNs). It has a wide range of applications in smart cities for building personalized scenarios. Current research typically uses sequential relationships to mine user preferences; however, it fails to sufficiently explore the spatial dependence of check-ins and the multi-perspective information they contain. To this end, this study proposes a Multiple Active Region Aware Network (MARAN), a novel routine-aware model for the next POI recommendation that simultaneously captures the user’s routine regularity and short-term preference changes from check-in records. The key to MARAN is its ability to decompose sophisticated user behavior into two parts. One is a stable routine part characterized by central-based graphs built from historical trajectories based on spatial aggregation. The other is an unstable preference part that obtains the user’s recent changes from short-term trajectories. Moreover, a neighborhood-aware negative sampler based on adjacent areas was designed to alleviate spatial sparsity, that is, the imbalance between positive and negative samples during model training. Experiments on two real-world datasets demonstrated that MARAN outperformed state-of-the-art methods.

RESUS: Warm-up Cold Users via Meta-learning Residual User Preferences in CTR Prediction

Click-through Rate (CTR) prediction on cold users is a challenging task in recommender systems. Recent researches have resorted to meta-learning to tackle the cold-user challenge, which either perform few-shot user representation learning or adopt optimization-based meta-learning. However, existing methods suffer from information loss or inefficient optimization process, and they fail to explicitly model global user preference knowledge, which is crucial to complement the sparse and insufficient preference information of cold users. In this article, we propose a novel and efficient approach named RESUS, which decouples the learning of global preference knowledge contributed by collective users from the learning of residual preferences for individual users. Specifically, we employ a shared predictor to infer basis user preferences, which acquires global preference knowledge from the interactions of different users. Meanwhile, we develop two efficient algorithms based on the nearest neighbor and ridge regression predictors, which infer residual user preferences via learning quickly from a few user-specific interactions. Extensive experiments on three public datasets demonstrate that our RESUS approach is efficient and effective in improving CTR prediction accuracy on cold users, compared with various state-of-the-art methods.

Rating Prediction based on Optimal Review Topics: A Proposed Latent Factors-Optimal Topics Hybrid Approach

Rating prediction is an inevitable problem which recommender systems (RS) need to address. Its goal is to accurately predict the rating a user will assign to a particular item. Predictions which utilize numerical ratings and review texts are biased and have low accuracy. Also, existing topic-based rating prediction approaches focus on finding the most preferred items through the identification of latent topics expressed in users’ review texts. Even though the latent topics seem to represent most user review texts, they do not necessarily capture each user’s preferences. The goal of this work is then to develop a more accurate model by considering product review texts analysis so as to gain additional preference knowledge. Hence, a hybrid algorithm that optimizes the latent topics is proposed. Specifically, the proposed approach finds appropriate weights for the topics of each review text. Rating prediction is critical task for RS because slight performance enhancement of the prediction accuracy results into significant improvements in recommendations. Experimental evaluation over real-world datasets revealed performance improvements of the proposed approach compared to alternative models. The proposed model can be used by RS in various domain such as e-learning, movie and hotel rating.

Meta-Path Based Neighbors for Behavioral Target Generalization in Sequential Recommendation

Click-through rate (CTR) prediction is a crucial task in recommender systems, which aims to model users’ dynamic preferences from their historical behaviors. To achieve this goal, most of the previous models adopt sequential neural networks (e.g., GRU) to encode the historical interactions into item representations for recommendations. Though these methods can perform well on recommending highly relevant items to users, we argue that such models are sub-optimal for the long-term user experience due to highly skewed recommendations: Monotonous items with similar subjects get more exposure because of inadequate interest explorations. Thus, some items which are not quite relevant to the users’ historical preferences should be considered. To address these limitations, we propose a Heterogeneous Graph Enhanced Sequential Neural Network, HGESNN, to explore the interests of users beyond their historical interactions by explicitly modeling item relations with meta-path constructions. We incorporate a transformer-based network to embed personalized user intents into sequential learning. In the experiments on both public and industrial datasets, HGESNN significantly outperforms the state-of-the-art solutions. Specifically, HGESNN has been deployed in the main traffic of our Image-Text feed recommender system, which obtains 6.28%, 6.82%, and 4.77% CTR gains on news, novels, and entertainment contents, respectively.

A CTR Prediction Model With Double Matrix-Level Cross-Features

CTR prediction is an important task in recommender systems, which is used to estimate the likelihood of a user clicking on an advertisement.In the past, the CTR prediction model based on the deep neural network mainly obtains the implicit feature combination of the model at the bit-wise level, and the interpretability and generalization of the model are poor. At the same time, the prediction accuracy of the model is poor. For the above problems, We propose a click-through rate prediction model (DTM) with double matrix-level cross-features. The model integrates various components such as multi-head self-attention, residual network and interaction network into an end-to-end model, and automatically obtains explicit feature combinations at the vector-wise level and bit-wise level, which not only has better interpretability, generalization and memory, and reduce the inherent flaws and engineering complexity of multi-modules. The experimental results show that on the datasets Criteo and Avazu, compared with other state-of-the-art CTR prediction models, the AUC values of the DTM model are increased by 4% and 3% on average, and the loss values are decreased by 3.5% and 2.8% on average, respectively.

A unified task recommendation strategy for realistic mobile crowdsourcing system

A well-designed task recommendation framework aims to protect the data quality as well as increase the task execution results. However, current crowdsourcing systems ignore the fact that there are few duplicate task expectations because of the budget limitation in realistic conditions. Besides, a practical crowdsourcing system needs to recommend new tasks without previous knowledge about the concrete task content due to short task lifespan. Thus, most of the existing studies are not applicable due to the idealized assumptions. In this paper, we formally define the problem and prove it is NP-Hard. For the problem, we design a unified task recommendation system for realistic conditions to address the mentioned problems, Pioneer-Assisted Task RecommendatiON (PATRON) framework. The framework first selects a set of pioneer workers to collect initial knowledge of the new tasks. Then it adopts the k-medoids clustering algorithm to split the workers into subsets based on the worker similarity. Cluster selection and worker pruning provides accurate and efficient recommendations that satisfy the valid recommendation requirements from requesters. Finally, we conducted our experiments based on real datasets from a famous Chinese crowdsourcing platform, Tencent SOHO. The experimental results show the efficiency and accuracy of PATRON compared with three baseline methods from several perspectives, such as recommendation success rate and recommended worker quality.

CoFi-points

With the explosive growth of web resources, an increasingly important task in recommender systems is to provide high-quality personalized services by learning users’ preferences from historically observed information. As an effective preference learning technology, collaborative filtering has been widely extended to model the one-class or implicit feedback data, which is known as one-class collaborative filtering (OCCF). For a long time, pairwise ranking-oriented learning scheme has been viewed as a superior solution than the pointwise scheme for OCCF due to its higher accuracy in most cases. However, we argue that with appropriate model design, pointwise preference learning can achieve comparable or even better performance than the counterpart, i.e., pairwise preference learning. In particular, we propose a new preference assumption, i.e., pointwise preference on user/item-set. Based on this new assumption, we develop a novel, simple, and flexible solution called collaborative filtering via pointwise preference learning on user/item-set (CoFi-points). Furthermore, we derive two specific algorithms of CoFi-points with respect to the involved user-set and item-set, i.e., CoFi-points(u) and CoFi-points(i), referring to preference assumptions defined on user-set and item-set, respectively. Finally, we conduct extensive empirical studies on four real-world datasets with the state-of-the-art methods, and find that our solution can achieve very promising performance with respect to several ranking-oriented evaluation metrics.

Intention Nets: Psychology-Inspired User Choice Behavior Modeling for Next-Basket Prediction

Human behaviors are complex, which are often observed as a sequence of heterogeneous actions. In this paper, we take user choices for shopping baskets as a typical case to study the complexity of user behaviors. Most of existing approaches often model user behaviors in a mechanical way, namely treating a user action sequence as homogeneous sequential data, such as hourly temperatures, which fails to consider the complexity in user behaviors. In fact, users' choices are driven by certain underlying intentions (e.g., feeding the baby or relieving pain) according to Psychological theories. Moreover, the durations of intentions to drive user actions are quite different; some of them may be persistent while others may be transient. According to Psychological theories, we develop a hierarchical framework to describe the goal, intentions and action sequences, based on which, we design Intention Nets (IntNet). In IntNet, multiple Action Chain Nets are constructed to model the user actions driven by different intentions, and a specially designed Persistent-Transient Intention Unit models the different intention durations. We apply the IntNet to next-basket prediction, a recent challenging task in recommender systems. Extensive experiments on real-world datasets show the superiority of our Psychology-inspired model IntNet over the state-of-the-art approaches.

Task recommender system using semantic clustering to identify the right personnel

PurposeText mining is growing in importance proportionate to the growth of unstructured data and its applications are increasing day by day from knowledge management to social media analysis. Mapping skillset of a candidate and requirements of job profile is crucial for conducting new recruitment as well as for performing internal task allocation in the organization. The automation in the process of selecting the candidates is essential to avoid bias or subjectivity, which may occur while shuffling through thousands of resumes and other informative documents. The system takes skillset in the form of documents to build the semantic space and then takes appraisals or resumes as input and suggests the persons appropriate to complete a task or job position and employees needing additional training. The purpose of this study is to extend the term-document matrix and achieve refined clusters to produce an improved recommendation. The study also focuses on achieving consistency in cluster quality in spite of increasing size of data set, to solve scalability issues.Design/methodology/approachIn this study, a synset-based document matrix construction method is proposed where semantically similar terms are grouped to reduce the dimension curse. An automated Task Recommendation System is proposed comprising synset-based feature extraction, iterative semantic clustering and mapping based on semantic similarity.FindingsThe first step in knowledge extraction from the unstructured textual data is converting it into structured form either as Term frequency–Inverse document frequency (TF-IDF) matrix or synset-based TF-IDF. Once in structured form, a range of mining algorithms from classification to clustering can be applied. The algorithm gives a better feature vector representation and improved cluster quality. The synset-based grouping and feature extraction for resume data optimizes the candidate selection process by reducing entropy and error and by improving precision and scalability.Research limitations/implicationsThe productivity of any organization gets enhanced by assigning tasks to employees with a right set of skills. Efficient recruitment and task allocation can not only improve productivity but also cater to satisfy employee aspiration and identifying training requirements.Practical implicationsIndustries can use the approach to support different processes related to human resource management such as promotions, recruitment and training and, thus, manage the talent pool.Social implicationsThe task recommender system creates knowledge by following the steps of the knowledge management cycle and this methodology can be adopted in other similar knowledge management applications.Originality/valueThe efficacy of the proposed approach and its enhancement is validated by carrying out experiments on the benchmarked dataset of resumes. The results are compared with existing techniques and show refined clusters. That is Absolute error is reduced by 30 per cent, precision is increased by 20 per cent and dimensions are lowered by 60 per cent than existing technique. Also, the proposed approach solves issue of scalability by producing improved recommendation for 1,000 resumes with reduced entropy.

Application of Transfer Learning in Task Recommendation System

In order to meet the need of individualized learning, the current network teaching platform all bring in recommendation system, including the recommendation of learning resources, the recommendation of task and exercise, etc. Problems like outdated datum or sparse datum still exist in teaching platform recommendation. Take task recommendation system of computer courses as an example, the computer knowledge updates quickly. The old knowledge will be out of date in two to three years. Another case is that, when the website was put into use, as the website traffic is quite low, the datum of task finished by the students are quite little. Both of those two cases can cause sparse datum of task system in network teaching platform. In order to solve this problem, this article tries to apply transfer learning into network teaching recommendation system, and verify its feasibility through experiment.

NNIA-RS: A multi-objective optimization based recommender system

With the rapid development of the Internet, we have entered an era of information explosion. In this kind of situation, recommender systems appear. It is meaningful that recommender systems help people to get useful and valuable items from massive data (e.g. movies, music, books, jokes). The traditional recommender systems especially collaborative filtering make recommendations based on similarity between items or users. Under this strategy, the similarities between the items in the recommendation list are extremely high so as to guarantee the accuracy of a recommendation, but it easily results in the loss of diversity. Actually, the task of recommender systems can be modeled as a multi-objective optimization problem. By taking into account of the recommendation accuracy and diversity, a multi-objective evolutionary algorithm for recommender systems is proposed. Each run of the proposed algorithm can produce a set of nondominated solutions. Each solution denotes a unique recommendation list to a target user. Experiments demonstrate that the proposed algorithm is promising and effective in terms of recommendation diversity and novelty.

A survey of collaborative filtering based social recommender systems

Recommendation plays an increasingly important role in our daily lives. Recommender systems automatically suggest to a user items that might be of interest to her. Recent studies demonstrate that information from social networks can be exploited to improve accuracy of recommendations. In this paper, we present a survey of collaborative filtering (CF) based social recommender systems. We provide a brief overview over the task of recommender systems and traditional approaches that do not use social network information. We then present how social network information can be adopted by recommender systems as additional input for improved accuracy. We classify CF-based social recommender systems into two categories: matrix factorization based social recommendation approaches and neighborhood based social recommendation approaches. For each category, we survey and compare several representative algorithms.

The Diffusion of a Task Recommendation System to Facilitate Contributions to an Online Community

This paper studies the diffusion of SuggestBot, an intelligent task recommendation system that helps people find articles to edit in Wikipedia. We investigate factors that predict who adopts SuggestBot and its impact on adopters' future contributions to this online community. Analyzing records of participants' activities in Wikipedia, we found that both individual characteristics and social ties influence adoption. Specially, we found that highly involved contributors were more likely to adopt SuggestBot; interpersonal exposure to innovation, cohesion, and tie homophily all substantially increased the likelihood of adoption. However, connections to prominent, high-status contributors did not influence adoption. Finally, although the SuggestBot innovation saw limited distribution, adopters made significantly more contributions to Wikipedia after adoption than nonadopter counterparts in the comparison group.