Preference Integration in Context-Aware Recommendation

Recommender systems have been attracting much attention from both academia and industry because of their ability to capture user interests and generate personalized item recommendations. As the life pace in contemporary society speeds up, traditional recommender systems are inevitably limited by their disconnected interaction styles and low adaptivity to users’ evolving demands. Consequently, conversational recommender systems emerge as a prospective research area, where an intelligent dialogue agent is integrated with a recommender system. Conversational recommender systems possess the ability to accurately understand end-users’ intent or request and generate human-like dialogue responses when performing recommendations. However, existing conversational recommender systems only allow the systems to ask users for more preference information, while users’ further questions and concerns about the recommended items (e.g., enquiring the location of a recommended restaurant) can hardly be addressed. Though the recent task-oriented dialogue systems allow for two-way communications, they are not easy to train because of their high dependence on human guidance in terms of user intent recognition and system response generation. Hence, to enable two-way human-machine communications and tackle the challenges brought by manually crafted rules, we propose Conversational Recommender System with Adversarial Learning (CRSAL), a novel end-to-end system to tackle the task of conversational recommendation. In CRSAL, we innovatively design a fully statistical dialogue state tracker coupled with a neural policy agent to precisely capture each user’s intent from limited dialogue data and generate conversational recommendation actions. We further develop an adversarial Actor-Critic reinforcement learning approach to adaptively refine the quality of generated system actions, thus ensuring coherent human-like dialogue responses. Extensive experiments on two benchmark datasets fully demonstrate the superiority of CRSAL on conversational recommendation tasks.

The software development platform is an increasingly expanding industry. It is growing steadily due to the active research and sharing of artificial intelligence and deep learning. Further, predicting users’ propensity in this huge community and recommending a new repository is beneficial for researchers and users. Despite this, only a few researches have been done on the recommendation system of such platforms. In this study, we propose a method to model extensive user data of an online community with a deep learning-based recommendation system. This study shows that a new repository can be effectively recommended based on the accumulated big data from the user. Moreover, this study is the first study of the sequential recommendation system that provides a new dataset of a software development platform, which is as large as the prevailing datasets. The experiments show that the proposed dataset can be practiced in various recommendation tasks.

Graph convolutional networks (GCN), aiming to learn meaningful representations for graph data, has been popularly used in recommender systems since user-item interactions can be represented by a bipartite graph. However, GCN often suffers from the over-smoothing issue when it goes deeper, which implies that long paths between users and items are not very useful to embedding learning in recommender systems. To deal with long paths, introducing user and item links is a promising way to shorten the average path between users and items. However, due to introducing the user and item linkages, the bipartite graph becomes a heterogeneous graph. To this end, we propose a multi-aspect heterogeneous GCN model, which is to power GCN towards recommendation tasks by introducing user-user and item-item links to build the heterogeneous user- item graph. Then, we decompose it into three subgraphs and keep the global heterogeneous graph to preserve the original structure information. Specifically, we apply GCN to learn user and item representations from both the subgraphs and the global graph. After that, we have multi-aspect attention to integrate the representations from GCN to obtain the final user and item representation. Extensive experiments on three real-world datasets show that the proposed approaches achieve significant improvements over state-of-the-art methods.

Centralized recommender systems can not resolve the contradiction between high recommendation quality and timely response, as well as that between limited recommendation range and ever rich information on the Web. Distributed recommender systems are expected to improve the recommendation quality while maintaining high performance. Large-scale distributed recommendation involves the coordination of various heterogeneous resources which are located on different nodes and need to be represented uniformly and organized normally into the resource space so that a semantically interactive environment can be formed. In this paper the recommendation task is defined as a knowledge based workflow and the knowledge grid is exploited as the platform for knowledge sharing and knowledge service which provides the functions of knowledge discovery, knowledge fusion and knowledge based workflow definition. The rationale of the knowledge grid based intelligent electronic recommender systems (KGBIECRS) is discussed and the service oriented architecture of the knowledge grid is presented. The knowledge grid depends on the semantic grid as the semantically interactive platform to intelligently coordinate the heterogeneous resources on the grid so that the recommendation task submitted by the knowledge grid as a knowledge based workflow can be performed intelligently and adaptively. How to implement the system is also discussed. Finally an example of travel recommendation is given to elaborate the recommendation process.

Video recordings are complex media types. When we watch a movie, we can effortlessly register a lot of details conveyed to us (by the author) through different multimedia channels, in particular, the audio and visual modalities. To date, majority of movie recommender systems use collaborative filtering (CF) models or content-based filtering (CBF) relying on metadata (e.g., editorial such as genre or wisdom of the crowd such as user-generated tags) at their core since they are human-generated and are assumed to cover the ‘content semantics’ of movies by a great degree. The information obtained from multimedia content and learning from muli-modal sources (e.g., audio, visual and metadata) on the other hand, offers the possibility of uncovering relationships between modalities and obtaining an in-depth understanding of natural phenomena occurring in a video. These discerning characteristics of heterogeneous feature sets meet users’ differing information needs. In the context of this Ph.D. thesis [9], which is briefly summarized in the current extended abstract, approaches to automated extraction of multimedia information from videos and their integration with video recommender systems have been elaborated, implemented, and analyzed. Variety of tasks related to movie recommendation using multimedia content have been studied. The results of this thesis can motivate the fact that recommender system research can benefit from knowledge in multimedia signal processing and machine learning established over the last decades for solving various recommendation tasks.

Some recommender systems (RS) utilize Linked Open Data (LOD) to enhance the item descriptions in the local database. However, these systems do not yet take full advantage of the potential of RDF data for personalized retrieval. The work describes the strengths of LOD repositories as well as the challenges of RDF processing for recommendation tasks. Against the background of these characteristics, a recommendation engine, called SKOSRecommender (SKOSRec), was developed. The system utilizes SKOS annotations to determine similar items and provides a graph-based query language for on-the-fly retrieval from SPARQL endpoints. This enables novel retrieval approaches. For instance, the SKOSRec language facilitates the representation of individual user preferences as query-based statements. Hence, it is possible to generate a user profile with the help of a SPARQL-like request (preference querying). Additionally, the language enables subquerying with recommendation results and the usage of graph-based query patterns to formulate powerful filter conditions for result lists (expressive constraint-based queries). Besides, the language allows flexible combinations of graph- and search-based query patterns (i.e., advanced recommendation requests). Examples of such requests are rollup retrieval patterns or cross-domain queries. The novel approaches were evaluated in a series of offline and online experiments in different domains (travel RS, multimedia RS and scientific publication retrieval). The results show that most of the developed methods improve the quality of existing recommendation methods. Effects predominantly occurred in the performance dimensions of recall, novelty, and diversity. The positive evaluation results demonstrate the effectiveness of the new methods. Thus, the work can contribute to the advancement of personalized search techniques, which can be applied for semantic retrieval in LOD repositories as well as for typical recommendation tasks.

A dual learning-based recommendation approach

In this article, we present PrefRec, a general framework for developing RS using Preference Mining and Preference Aggregation techniques. We focus on Pairwise Preference Mining techniques allowing to predict which, between two objects, is the preferred one.A preliminary empirical study for analyzing the influence of the different factors involved in each of the five modules of PrefRec is presented.Keywordsrecommender systemspreference miningcontextual preferencespreference aggregation

It is meaningful for researchers to find the interested and high quality new papers. We propose the Joint Text and Influence Embedding recommendation model (JTIE) to consider both the paper quality and the content correlation. We train a paper embedding based on its core elements: contents, authors and publication venues. The quality of a new paper is evaluated based on the author authority and the venue reputation. The citation relationships between papers are considered asymmetric such that they can reflect the user’s consideration on the intrinsic influence of a paper. We learn user interests by one’s historical references or a set of query keywords. Finally, papers are recommended according to the relatedness between user interests and paper embeddings. We perform experiments against three real-world datasets. The results show that our model outperforms baseline methods on both the personalized recommendation and the query keywords based retrieval.

The point-of-interest (POI) recommender exploits check-in information from location-based social networks (LBSNs) to recommend POIs that match user preferences. The preferences of users vary with region/locality, consumption type, and co-consumers. For instance, the places preferred for friends may be different from the places preferred for family, and the places preferred in one locality may be different from the places in another locality. These dynamic preferences can be crucial for an efficient recommender system. A locality consists of different preference trends, such as “known for food and recreational sites.” In real-world, different sets of users might be attracted toward different preference trends, and some users' preferences might overlap across multiple preference sets. Hence, an efficient aggregation of locality preferences is essential for the recommender systems. Many existing studies simply group items by their category and use simple collaborative filtering (CF) for recommendations, however, such techniques cannot efficiently handle the aggregated locality preferences. We propose a hierarchical recommendation model termed Hierarchical Contextual Location Recommendation System (HiRecS) that formulates users' preferences as a hierarchical structure and models the locality trend using aggregated hierarchy. For a locality, the root of hierarchy contains preferred k items from a set of visitors, and the subsequent levels contain preference wise subsets of those items. We also present a hierarchy aggregation technique to aggregate the hierarchical preferences from a similar set of users. The aggregated hierarchy is then contextually exploited for POI sequence recommendation. The core contributions of this article are: 1) it formulates the locality trends as hierarchical structures, presents a hierarchy aggregation technique, and models the personalized POI preferences with the aggregated hierarchy; 2) it exploits the aggregated popular trends to generate contextual POI sequence recommendation; and 3) it extensively evaluates the proposed model with two real-world data sets and demonstrates a significant performance gain of 0.006-5.91 on diversity metrics, 0.0349-17.51 on displacement metrics, and 0.114-0.289 on normalized discounted cumulative gain (NDCG) metrics, when compared to several baselines and relevant studies.

The main question to address in this paper is to recommend optimal signal timing plans in real time under incidents by incorporating domain knowledge developed with the traffic signal timing plans tuned for possible incidents, and learning from historical data of both traffic and implemented signals timing. The effectiveness of traffic incident management is often limited by the late response time and excessive workload of traffic operators. This paper proposes a novel decision-making framework that learns from both data and domain knowledge to real-time recommend contingency signal plans that accommodate non-recurrent traffic, with the outputs from real-time traffic prediction at least 30 min in advance. Specifically, considering the rare occurrences of engagement of contingency signal plans for incidents, it is proposed to decompose the end-to-end recommendation task into two hierarchical models—real-time traffic prediction and plan association. The connections between the two models are learnt through metric learning, which reinforces partial-order preferences observed from historical signal engagement records. The effectiveness of this approach is demonstrated by testing this framework on the traffic network in Cranberry Township, Pennsylvania, U.S., in 2019. Results show that the recommendation system has a precision score of 96.75% and recall of 87.5% on the testing plan, and makes recommendations an average of 22.5 min lead time ahead of Waze alerts. The results suggest that this framework is capable of giving traffic operators a significant time window to access the conditions and respond appropriately.

Due to the complementary nature of graph neural networks and structured data in recommendations, recommendation systems using graph neural network techniques have become mainstream. However, there are still problems, such as sparse supervised signals and interaction noise, in the recommendation task. Therefore, this paper proposes a self-supervised graph attention collaborative filtering for recommendation (SGACF). The correlation between adjacent nodes is deeply mined using a multi-head graph attention network to obtain accurate node representations. It is worth noting that self-supervised learning is brought in as an auxiliary task in the recommendation, where the supervision task is the main task. It assists model training for supervised tasks. A multi-view of a node is generated by the graph data-augmentation method. We maximize the consistency between its different views compared to the views of the same node and minimize the consistency between its different views compared to the views of other nodes. In this paper, the effectiveness of the method is illustrated by abundant experiments on three public datasets. The results show its significant improvement in the accuracy of the long-tail item recommendation and the robustness of the model.

Cold start problem is one of the most challenging and long-standing problems in recommender systems, and cross-domain recommendation (CDR) methods are effective for tackling it. Most cold-start related CDR methods require training a mapping function between high-dimensional embedding space using overlapping user data. However, the overlapping data is scarce in many recommendation tasks, which makes it difficult to train the mapping function. In this paper, we propose a new approach for CDR, which aims to alleviate the training difficulty. The proposed method can be viewed as a special parameterization of the mapping function without hurting expressiveness, which makes use of non-overlapping user data and leads to effective optimization. Extensive experiments on two real-world CDR tasks are performed to evaluate the proposed method. In the case that there are few overlapping data, the proposed method outperforms the existed state-of-the-art method by 14% (relative improvement).

Recommender systems play a pivotal role in providing relevant content to users.With the rapid development of large language models (LLMs), researchers have begun utilizing LLMs to build more powerful recommender systems.However, existing approaches that focus on aligning LLMs with recommendation tasks do not fully leverage their sequential information processing capabilities, leading to suboptimal performance.In this paper, we propose a novel system called compressed vocabulary expansion (CoVE).In CoVE, each item is assigned a unique ID within the expanded vocabulary.Our framework effectively capitalizes on sequence understanding abilities of LLMs, significantly enhancing their performance on recommendation tasks.Additionally, we compress the embedding layer, making CoVE practical for largescale industrial applications.The effectiveness and performance of CoVE are demonstrated through comprehensive experiments on multiple recommendation datasets and comparisons with prior works.

With the rapid development of recommendation systems, models and algorithms supporting the core of recommendation systems have emerged one after another, and researchers have attempted to optimize them. However, the structure of these models is complex. Popular deep neural networks often achieve the highest utilization of data by increasing the number of hidden layers, ignoring the problems of exploding and vanishing gradients and even the entire degradation of the networks. However, researchers pay too much attention to algorithms and models and do not consider the dataset itself. Methods for processing data and finding possible connections between the data and models have become new explorable points. Cold start is also a problem that researchers have been trying to solve and optimize since the birth of the recommendation system. Recent studies also provide good ideas for solving cold start, but the problem is that researchers still do not focus on datasets. In order to fill the gap in the exploration and research of datasets, this paper takes the long tail distribution and cold start problems that are common in recommendation systems such as the starting point, combines the residual network in computer vision with deep learning, and proposes the aging mechanism of datasets. In this paper, a multi-layer residual network based on aging mechanisms called Aging Residual Factorization Machines (ARFM) is proposed. Parallel experiments with other model algorithms are carried out on three datasets of different sizes and categories. Experimental results show that ARFM achieve performance advantages under the premise of different recommendation tasks.