A user-centric evaluation of context-aware recommendations for a mobile news service

Traditional recommender systems provide personal suggestions based on the user’s preferences, without taking into account any additional contextual information, such as time or device type. The added value of contextual information for the recommendation process is highly dependent on the application domain, the type of contextual information, and variations in users’ usage behavior in different contextual situations. This paper investigates whether users utilize a mobile news service in different contextual situations and whether the context has an influence on their consumption behavior. Furthermore, the importance of context for the recommendation process is investigated by comparing the user satisfaction with recommendations based on an explicit static profile, content-based recommendations using the actual user behavior but ignoring the context, and context-aware content-based recommendations incorporating user behavior as well as context. Considering the recommendations based on the static profile as a reference condition, the results indicate a significant improvement for recommendations that are based on the actual user behavior. This improvement is due to the discrepancy between explicitly stated preferences (initial profile) and the actual consumption behavior of the user. The context-aware content-based recommendations did not significantly outperform the content-based recommendations in our user study. Context-aware content-based recommendations may induce a higher user satisfaction after a longer period of service operation, enabling the recommender to overcome the cold-start problem and distinguish user preferences in various contextual situations.     

标签推荐方法研究综述

随着互联网信息的爆炸式增长,标签(由用户指定用来描述项目的关键词)在互联网信息检索领域中变得越来越重要.为在线内容赋予合适的标签,有利于更高效的内容组织和内容消费.而标签推荐通过辅助用户进行打标签的操作,极大地提升了标签的质量,标签推荐也因此受到了研究者们的广泛关注.总结出标签推荐任务的三大特性,即项目内容的多样性、标...     

Inferring contextual preferences using deep encoder-decoder learners

ABSTRACT

Context-aware systems enable the sensing and analysis of user context in order to provide personalised services. Our study is part of growing research efforts examining how high-dimensional data collected from mobile devices can be utilised to infer users’ dynamic preferences that are learned over time. We suggest novel methods for inferring the category of the item liked in a specific contextual situation, by applying encoder-decoder learners (long short-term memory networks and auto encoders) on mobile sensor data. In these approaches, the encoder-decoder learners reduce the dimensionality of the contextual features to a latent representation which is learned over time. Given new contextual sensor data from a user, the latent patterns discovered from each deep learner is used to predict the liked item’s category in the given context. This can greatly enhance a variety of services, such as mobile online advertising and context-aware recommender systems. We demonstrate our contribution with a point of interest (POI) recommender system in which we label contextual situations with the items’ categories. Empirical results utilising a real world data set of contextual situations derived from mobile phones sensors log show a significant improvement (up to 73% improvement) in prediction accuracy compared with state of the art classification methods.     

Modeling context-aware and intention-aware in-car infotainment systems

It is fundamental to understand users’ intentions to support them when operating a computer system with a dynamically varying set of functions, e.g., within an in-car infotainment system. The system needs to have sufficient information about its own and the user’s context to predict those intentions. Although the development of current in-car infotainment systems is already model-based, explicitly gathering and modeling contextual information and user intentions is currently not supported. However, manually creating software that understands the current context and predicts user intentions is complex, error-prone and expensive. Model-based development can help in overcoming these issues. In this paper, we present an approach for modeling a user’s intention based on Bayesian networks. We support developers of in-car infotainment systems by providing means to model possible user intentions according to the current context. We further allow modeling of user preferences and show how the modeled intentions may change during run-time as a result of the user’s behavior. We demonstrate feasibility of our approach using an industrial case study of an intention-aware in-car infotainment system. Finally, we show how modeling of contextual information and modeling user intentions can be combined by using model transformation.     

Context relevance assessment and exploitation in mobile recommender systems

In order to generate relevant recommendations, a context-aware recommender system (CARS) not only makes use of user preferences, but also exploits information about the specific contextual situation in which the recommended item will be consumed. For instance, when recommending a holiday destination, a CARS could take into account whether the trip will happen in summer or winter. It is unclear, however, which contextual factors are important and to which degree they influence user ratings. A large amount of data and complex context-aware predictive models must be exploited to understand these relationships. In this paper, we take a new approach for assessing and modeling the relationship between contextual factors and item ratings. Rather than using the traditional approach to data collection, where recommendations are rated with respect to real situations as participants go about their lives as normal, we simulate contextual situations to more easily capture data regarding how the context influences user ratings. To this end, we have designed a methodology whereby users are asked to judge whether a contextual factor (e.g., season) influences the rating given a certain contextual condition (e.g., season is summer). Based on the analyses of these data, we built a context-aware mobile recommender system that utilizes the contextual factors shown to be important. In a subsequent user evaluation, this system was preferred to a similar variant that did not exploit contextual information.     

RecAm: a collaborative context-aware framework for multimedia recommendations in an ambient intelligence environment

With an ever-increasing accessibility to different multimedia contents in real-time, it is difficult for users to identify the proper resources from such a vast number of choices. By utilizing the user’s context while consuming diverse multimedia contents, we can identify different personal preferences and settings. However, there is a need to reinforce the recommendation process in a systematic way, with context-adaptive information. The contributions of this paper are twofold. First, we propose a framework, called RecAm, which enables the collection of contextual information and the delivery of resulted recommendation by adapting the user’s environment using Ambient Intelligent (AmI) Interfaces. Second, we propose a recommendation model that establishes a bridge between the multimedia resources, user joint preferences, and the detected contextual information. Hence, we obtain a comprehensive view of the user’s context, as well as provide a personalized environment to deliver the feedback. We demonstrate the feasibility of RecAm with two prototypes applications that use contextual information for recommendations. The offline experiment conducted shows the improvement of delivering personalized recommendations based on the user’s context on two real-world datasets.     

A novel framework to alleviate the sparsity problem in context-aware recommender systems

Recommender systems have become indispensable for services in the era of big data. To improve accuracy and satisfaction, context-aware recommender systems (CARSs) attempt to incorporate contextual information into recommendations. Typically, valid and influential contexts are determined in advance by domain experts or feature selection approaches. Most studies have focused on utilizing the unitary context due to the differences between various contexts. Meanwhile, multi-dimensional contexts will aggravate the sparsity problem, which means that the user preference matrix would become extremely sparse. Consequently, there are not enough or even no preferences in most multi-dimensional conditions. In this paper, we propose a novel framework to alleviate the sparsity issue for CARSs, especially when multi-dimensional contextual variables are adopted. Motivated by the intuition that the overall preferences tend to show similarities among specific groups of users and conditions, we first explore to construct one contextual profile for each contextual condition. In order to further identify those user and context subgroups automatically and simultaneously, we apply a co-clustering algorithm. Furthermore, we expand user preferences in a given contextual condition with the identified user and context clusters. Finally, we perform recommendations based on expanded preferences. Extensive experiments demonstrate the effectiveness of the proposed framework.     

A Framework for Collaborative, Content-Based and Demographic Filtering

We discuss learning a profile of user interests for recommending information sources such as Web pages or news articles. We describe the types of information available to determine whether to recommend a particular page to a particular user. This information includes the content of the page, the ratings of the user on other pages and the contents of these pages, the ratings given to that page by other users and the ratings of these other users on other pages and demographic information about users. We describe how each type of information may be used individually and then discuss an approach to combining recommendations from multiple sources. We illustrate each approach and the combined approach in the context of recommending restaurants.     

Large-scale user modeling with recurrent neural networks for music discovery on multiple time scales

The amount of content on online music streaming platforms is immense, and most users only access a tiny fraction of this content. Recommender systems are the application of choice to open up the collection to these users. Collaborative filtering has the disadvantage that it relies on explicit ratings, which are often unavailable, and generally disregards the temporal nature of music consumption. On the other hand, item co-occurrence algorithms, such as the recently introduced word2vec-based recommenders, are typically left without an effective user representation. In this paper, we present a new approach to model users through recurrent neural networks by sequentially processing consumed items, represented by any type of embeddings and other context features. This way we obtain semantically rich user representations, which capture a user’s musical taste over time. Our experimental analysis on large-scale user data shows that our model can be used to predict future songs a user will likely listen to, both in the short and long term.

     

Non-dominated differential context modeling for context-aware recommendations

Context plays an important role in the process of decision making. A user’s preferences on the items may vary from contexts to contexts, e.g., a user may prefer to watch a different type of the movies, if he or she is going to enjoy the movie with partner rather than with children. Context-aware recommender systems, therefore, were developed to adapt the recommendations to different contextual situations, such as time, location, companion, etc. Differential context modeling is a series of recommendation models which incorporate contextual hybrid filtering into the neighborhood based collaborative filtering approaches. In this paper, we propose to enhance differential context modeling by utilizing a non-dominated user neighborhood. The notion of dominance relation was originally proposed in multi-objective optimization, and it was reused to definite non-dominated user neighborhood in collaborative filtering recently. These non-dominated user neighbors refer to the neighbors that dominate others from different perspectives of the user similarities, such as the user-user similarities based on ratings, demographic information, social relationships, and so forth. In this paper, we propose to identify the non-dominated user neighborhood by exploiting user-user similarities over multiple contextual preferences. Our experimental results can demonstrate the effectiveness of the proposed approaches in comparison with popular context-aware collaborative filtering models over five real-world contextual rating data sets.

     

一种上下文移动用户偏好自适应学习方法

针对移动网络对个性化移动网络服务系统的性能提出了更高的要求,但现有研究难以自适应地修改上下文移动用户偏好以为移动用户提供实时、准确的个性化移动网络服务的问题,提出了一种上下文移动用户偏好自适应学习方法,在保证精确度的基础上缩短了学习的响应时间.首先,通过分析移动用户行为日志来判断移动用户行为是否受上下文影响,并在此基础上判断移动用户行为是否发生变化.然后,根据判断结果对上下文移动用户偏好进行修正.在对发生变化的上下文移动用户偏好进行学习时,将上下文引入到最小二乘支持向量机中,进一步提出了基于上下文最小二乘支持向量机(C-LSSVM)的上下文移动用户偏好学习方法.最后,实验结果表明,当综合考虑精确度和响应时间两方面因素时,所提出的方法优于其他学习方法,并且可应用于个性化移动网络服务系统中.     

Search personalization through query and page topical analysis

Thousands of users issue keyword queries to the Web search engines to find information on a number of topics. Since the users may have diverse backgrounds and may have different expectations for a given query, some search engines try to personalize their results to better match the overall interests of an individual user. This task involves two great challenges. First the search engines need to be able to effectively identify the user interests and build a profile for every individual user. Second, once such a profile is available, the search engines need to rank the results in a way that matches the interests of a given user. In this article, we present our work towards a personalized Web search engine and we discuss how we addressed each of these challenges. Since users are typically not willing to provide information on their personal preferences, for the first challenge, we attempt to determine such preferences by examining the click history of each user. In particular, we leverage a topical ontology for estimating a user’s topic preferences based on her past searches, i.e. previously issued queries and pages visited for those queries. We then explore the semantic similarity between the user’s current query and the query-matching pages, in order to identify the user’s current topic preference. For the second challenge, we have developed a ranking function that uses the learned past and current topic preferences in order to rank the search results to better match the preferences of a given user. Our experimental evaluation on the Google query-stream of human subjects over a period of 1 month shows that user preferences can be learned accurately through the use of our topical ontology and that our ranking function which takes into account the learned user preferences yields significant improvements in the quality of the search results.     

Personality,User Preferences and Behavior in Recommender systems

This paper reports on a study of 1840 users of the MovieLens recommender system with identified Big-5 personality types. Based on prior literature that suggests that personality type is a stable predictor of user preferences and behavior, we examine factors of user retention and engagement, content preferences, and rating patterns to identify recommender-system related behaviors and preferences that correlate with user personality. We find that personality traits correlate significantly with behaviors and preferences such as newcomer retention, intensity of engagement, activity types, item categories, consumption versus contribution, and rating patterns.     

Contextual text/non-text stroke classification in online handwritten notes with conditional random fields

Analysing online handwritten notes is a challenging problem because of the content heterogeneity and the lack of prior knowledge, as users are free to compose documents that mix text, drawings, tables or diagrams. The task of separating text from non-text strokes is of crucial importance towards automated interpretation and indexing of these documents, but solving this problem requires a careful modelling of contextual information, such as the spatial and temporal relationships between strokes. In this work, we present a comprehensive study of contextual information modelling for text/non-text stroke classification in online handwritten documents. Formulating the problem with a conditional random field permits to integrate and combine multiple sources of context, such as several types of spatial and temporal interactions. Experimental results on a publicly available database of freely hand-drawn documents demonstrate the superiority of our approach and the benefit of contextual information combination for solving text/non-text classification.     

对话推荐算法研究综述

推荐系统是一种通过理解用户的兴趣和偏好帮助用户过滤大量无效信息并获取感兴趣的信息或者物品的信息过滤系统.目前主流的推荐系统主要基于离线的、历史的用户数据，不断训练和优化线下模型，继而为在线的用户推荐物品，这类训练方式主要存在3个问题：基于稀疏且具有噪声的历史数据估计用户偏好的不可靠估计、对影响用户行为的在线上下文环境因素的忽略和默认用户清楚自身偏好的不可靠假设.由于对话系统关注于用户的实时反馈数据，获取用户当前交互的意图，因此“对话推荐”通过结合对话形式与推荐任务成为解决传统推荐问题的有效手段.对话推荐将对话系统实时交互的数据获取方式应用到推荐系统中，采用了与传统推荐系统不同的推荐思路，通过利用在线交互信息，引导和捕捉用户当前的偏好兴趣，并及时进行反馈和更新.在过去的几年里，越来越多的研究者开始关注对话推荐系统，这一方面归功于自然语言处理领域中语音助手以及聊天机器人技术的广泛使用，另一方面受益于强化学习、知识图谱等技术在推荐策略中的成熟应用.将对话推荐系统的整体框架进行梳理，将对话推荐算法研究所使用的数据集进行分类，同时对评价对话推荐效果的相关指标进行讨论，重点关注于对话推荐系统中的后台对话策略与推荐逻辑，对近年来的对话推荐算法进行综述，最后对对话推荐领域的未来发展方向进行展望.     

Learning to embed music and metadata for context-aware music recommendation

Contextual factors greatly influence users’ musical preferences, so they are beneficial remarkably to music recommendation and retrieval tasks. However, it still needs to be studied how to obtain and utilize the contextual information. In this paper, we propose a context-aware music recommendation approach, which can recommend music pieces appropriate for users’ contextual preferences for music. In analogy to matrix factorization methods for collaborative filtering, the proposed approach does not require music pieces to be represented by features ahead, but it can learn the representations from users’ historical listening records. Specifically, the proposed approach first learns music pieces’ embeddings (feature vectors in low-dimension continuous space) from music listening records and corresponding metadata. Then it infers and models users’ global and contextual preferences for music from their listening records with the learned embeddings. Finally, it recommends appropriate music pieces according to the target user’s preferences to satisfy her/his real-time requirements. Experimental evaluations on a real-world dataset show that the proposed approach outperforms baseline methods in terms of precision, recall, F1 score, and hitrate. Especially, our approach has better performance on sparse datasets.     

Impact of query intent and search context on clickthrough behavior in sponsored search

Implicit feedback techniques may be used for query intent detection, taking advantage of user behavior to understand their interests and preferences. In sponsored search, a primary concern is the user’s interest in purchasing or utilizing a commercial service, or what is called online commercial intent. In this paper, we develop a methodology for employing the content of search engine result pages (SERPs), along with the information obtained from query strings, to study characteristics of query intent, with a particular focus on sponsored search. Our work represents a step toward the development and evaluation of an ontology for commercial search, considering queries that reference specific products, brands, and retailers. Characteristics of query categories are studied with respect to aggregated user clickthrough behavior on advertising links. We present a model for clickthrough behavior that considers the influence of such factors as the location of ads and the rank of ads, along with query category. We evaluate our work using a large corpus of clickthrough data obtained from a major commercial search engine. In addition, the impact of query intent is studied on clickthrough rate, where a baseline model and the query intent model are compared for the purpose of calculating an expected ad clickthrough rate. Our findings suggest that query-based features, along with the content of SERPs, are effective in detecting query intent. Factors such as query category, the rank of an ad, and the total number of ads displayed on a result page relate to the context of the ad, rather than its content. We demonstrate that these context-related factors can have a major influence on expected clickthrough rate, suggesting that these factors should be taken into consideration when the performance of an ad is evaluated.     

Managing contextual preferences

To handle the overwhelming amount of information currently available, personalization systems allow users to specify through preferences which pieces of data interest them. Most often, users have different preferences depending on context. In this paper, we introduce a model for expressing such contextual preferences. Context is modeled using a set of hierarchical attributes, thus allowing context specification at various levels of detail. We formulate the context resolution problem as the problem of selecting appropriate preferences based on context for personalizing a query. We also propose algorithms for context resolution based on data structures that index preferences by exploiting the hierarchical nature of the context attributes. Finally, we evaluate our approach from two perspectives: usability and performance. Usability evaluates the overheads imposed on users for specifying context-dependent preferences, as well as their satisfaction from the quality of the results. Our performance results focus on the context resolution using the proposed indexes.     

Protecting query privacy in location-based services

The popularity of location-based services (LBSs) leads to severe concerns on users’ privacy. With the fast growth of Internet applications such as online social networks, more user information becomes available to the attackers, which allows them to construct new contextual information. This gives rise to new challenges for user privacy protection and often requires improvements on the existing privacy-preserving methods. In this paper, we classify contextual information related to LBS query privacy and focus on two types of contexts—user profiles and query dependency: user profiles have not been deeply studied in LBS query privacy protection, while we are the first to show the impact of query dependency on users’ query privacy. More specifically, we present a general framework to enable the attackers to compute a distribution on users with respect to issuing an observed request. The framework can model attackers with different contextual information. We take user profiles and query dependency as examples to illustrate the implementation of the framework and their impact on users’ query privacy. Our framework subsequently allows us to show the insufficiency of existing query privacy metrics, e.g., k-anonymity, and propose several new metrics. In the end, we develop new generalisation algorithms to compute regions satisfying users’ privacy requirements expressed in these metrics. By experiments, our metrics and algorithms are shown to be effective and efficient for practical usage.     

Context-aware recommendations through context and activity recognition in a mobile environment

The mobile Internet introduces new opportunities to gain insight in the user’s environment, behavior, and activity. This contextual information can be used as an additional information source to improve traditional recommendation algorithms. This paper describes a framework to detect the current context and activity of the user by analyzing data retrieved from different sensors available on mobile devices. The framework can easily be extended to detect custom activities and is built in a generic way to ensure easy integration with other applications. On top of this framework, a recommender system is built to provide users a personalized content offer, consisting of relevant information such as points-of-interest, train schedules, and touristic info, based on the user’s current context. An evaluation of the recommender system and the underlying context recognition framework shows that power consumption and data traffic is still within an acceptable range. Users who tested the recommender system via the mobile application confirmed the usability and liked to use it. The recommendations are assessed as effective and help them to discover new places and interesting information.