Item recommendation based on context-aware model for personalized u-healthcare service

A personalized service in the ubiquitous environment is to provide services or items, which reflect personal tastes, attitudes, and contexts. It is impossible to reflect the context information generated in u-healthcare environments due to the existing recommendation system performing the recommendation using the information directly input by users and application usage record only. This study develops a context-aware model using the context information provided by the context information model. The study applies it to the extraction of the missing value in a collaborative filtering process. The context-aware model reflects the information that selects items by users according to the appropriate context using the C-HMM and provides it to users. The solution of the missing value in the preference significantly affects the recommendation accuracy in a preference based item supply method. Thus, this study developed a new collaborative filtering for ubiquitous environments by reflecting the missing preference value and reflecting it to the collaborative filtering using the context-aware model. Also, the validity of this method will be evaluated by applying it to menu services in u-healthcare services.     

FCT: a fully-distributed context-aware trust model for location based service recommendation

With the popularity of location based service (LBS), a vast number of trust models for LBS recommendation (LBSR) have been proposed. These trust models are centralized in essence, and the trusted third party may collude with malicious service providers or cause the single-point failure problem. This work improves the classic certified reputation (CR) model and proposes a novel fully-distributed context-aware trust (FCT) model for LBSR. Recommendation operations are conducted by service providers directly and the trusted third party is no longer required in our FCT model. Besides, our FCT model also supports the movements of service providers due to its self-certified characteristic. Moreover, for easing the collusion attack and value imbalance attack, we comprehensively consider four kinds of factor weights, namely number, time decay, preference and context weights. Finally, a fully-distributed service recommendation scenario is deployed, and comprehensive experiments and analysis are conducted. The results indicate that our FCT model significantly outperforms the CR model in terms of the robustness against the collusion attack and value imbalance attack, as well as the service recommendation performance in improving the successful trading rates of honest service providers and reducing the risks of trading with malicious service providers.     

A multimodel context-aware tourism recommendation service: Approach and architecture

Recommender systems and services are now widely used to support decision-making in the fields characterized by the selection from a large number of alternatives with a significant influence of subjective preferences. A comprehensive multimodel approach to the development of context-aware recommender systems in the field of tourism information support is proposed. In particular, it is proposed to construct a recommender system based on loosely coupled modules, in which both personalized and nonpersonalized recommendation methods are implemented, and the synthesis module, which adapts the module system to the specific conditions of different kinds of initial information.     

面向不一致用户评价准则的在线服务推荐

客观上,用户的评价准则是由主观意识决定的,用户之间的评价准则不同导致多个用户对同一服务的评分不具备可比较性,不考虑不同用户评分的不可比较性所获得的服务推荐将难以满足用户个性偏好及其真实需求。为此,提出一种面向不一致用户评价准则的在线服务推荐方法,考虑用户偏好不一致时用户对在线服务的偏好关系,以偏好关系计算用户之间的相似度,并以此获得在线服务推荐结果。首先以用户-服务评分矩阵为基础建立用户对服务的偏好关系,其次根据偏好关系计算用户之间的相似度,然后以用户相似度为基础对用户未评分的服务进行评分预测,最后以预测评分的排序结果作为推荐结果。与经典的协同过滤推荐方法的比较实验,验证了本方法的有效性。实验表明,本方法获得的推荐结果能满足大多数用户的服务偏好,同时获得了比经典的协同过滤推荐方法更好的准确率。     

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.     

移动电话内容服务系统的个性化推荐

移动电话内容服务系统允许移动用户通过移动互联技术浏览、购买和下载系统内容,是当前移动增值领域研究的热点。具有较强的时空灵活性,但在信息浏览、查找方面存在明显的局限性。提出了一个基于移动电话内容服务系统的个性化推荐系统．介绍了从寻找目标用户到实现推荐的全过程。实验结果表明。所介绍的个性化推荐系统可以有助于解决内容服务系统用户访问受限、资源迷茫的问题。     

Agent-based architecture for context-aware and personalized event recommendation

In order to offer context-aware and personalized information, intelligent processing techniques are necessary. Different initiatives considering many contexts have been proposed, but users preferences need to be learned to offer contextualized and personalized services, products or information. Therefore, this paper proposes an agent-based architecture for context-aware and personalized event recommendation based on ontology and the spreading algorithm. The use of ontology allows to define the domain knowledge model, while the spreading activation algorithm learns user patterns by discovering user interests. The proposed agent-based architecture was validated with the modeling and implementation of eAgora? application, which was illustrated at the pervasive university context.     

The DYNAMOS approach to support context-aware service provisioning in mobile environments

To efficiently make use of information and services available in ubiquitous environments, mobile users need novel means for locating relevant content, where relevance has a user-specific definition. In the DYNAMOS project, we have investigated a hybrid approach that enhances context-aware service provisioning with peer-to-peer social functionalities. We have designed and implemented a system platform and application prototype running on smart phones to support this novel conception of service provisioning. To assess the feasibility of our approach in a real-world scenario, we conducted field trials in which the research subject was a community of recreational boaters.     

Web数据挖掘在个性化推荐服务的应用

随着Internet的迅速发展和普及,人们在网络上开展的活动也越来越多,尤其是电子商务的兴起,使得网上购物逐渐成为一种时尚.然而由于互联网上的信息量日益增长,人们面对这些庞大的信息感到无所适从,花费大量时间在搜寻信息上,因此如何为用户提供更好的服务变得非常迫切.本文以数据挖掘理论为基础,分析和探讨了关联规则挖掘算法并应用该方法实现web个性化推荐服务.     

u-BabSang: a context-aware food recommendation system

In this paper, we propose a context-aware food recommendation system for well-being care applications. The proposed system, called u-BabSang, provides individualized food recommendation lists at the dining table, and is based dietary advice in the typical Korean medical text. Our proposed system receives a user’s profile, physiological signals, and environmental information around the dining table in real time. To operate our system, we present a method for user specified analysis, and also describe time-division layered context integration which integrates the multiple contexts obtained from the sensors. Thus, our system recommends appropriate foods for each individual’s health at the table in real time.     

Contextualized mobile recommendation service based on interactive social network discovered from mobile users

Personal context is the most significant information for providing contextualized mobile recommendation services at a certain time and place. However, it is very difficult for service providers to be aware of the personal contexts, because each person’s activities and preferences are very ambiguous and depending on numerous unknown factors. In order to deal with this problem, we have focused on discovering social relationships (e.g., family, friends, colleagues and so on) between people. We have assumed that the personal context of a certain person is interrelated with those of other people, and investigated how to employ his neighbor’s contexts, which possibly have a meaningful influence on his personal context. It indicates that we have to discover implicit social networks which express the contextual dependencies between people. Thereby, in this paper, we propose an interactive approach to build meaningful social networks by interacting with human experts. Given a certain social relation (e.g., isFatherOf), this proposed systems can evaluate a set of conditions (which are represented as propositional axioms) asserted from the human experts, and show them a social network resulted from data mining tools. More importantly, social network ontology has been exploited to consistently guide them by proving whether the conditions are logically verified, and to refine the discovered social networks. We expect these social network is applicable to generate context-based recommendation services. In this research project, we have applied the proposed system to discover the social networks between mobile users by collecting a dataset from about two millions of users.     

Modular framework support for context-aware mobile cinema

In this paper we describe a novel approach for interactive cinema based on context-aware narration using handheld computers. The paper describes both the artistic approach and the ubiquitous computing framework developed to realize the scenario. This framework has been used in various projects, including the described video production course at the ISNM, where five interactive cinema concepts have been developed and shown during a public demonstration. In our approach, a new type of user experience has been established by placing the viewer inside the movie’s physical locations during playback. Moreover, the developed ubiquitous computing framework provides a foundation for future work in the area of ad-hoc, service-oriented Ubicomp scenarios.     

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.     

An application framework for mobile,context-aware trails

In this paper we describe the design, implementation and evaluation of a software framework that supports the development of mobile, context-aware trails-based applications. A trail is a contextually scheduled collection of activities and represents a generic model that can be used to satisfy the activity management requirements of a wide range of context-based time management applications. Trails overcome limitations with traditional time management techniques based on static to-do lists by dynamically reordering activities based on emergent context.     

Emotional index measurement method for context-aware service

Emotion can be key context information when providing individualized services. Although some physical sensors can acquire emotional state, installing them in all spaces would be too cost-prohibitive—aside from raising strong privacy concerns in exposing individuals’ physical conditions. These issues lead us to investigate other ways to estimate users’ emotions. In particular, we focus on estimating degrees of emotion from users’ writing and speech. In this paper, we propose an emotion estimation methodology for context-aware services. To do so, we developed a method for calculating the degree of emotion in emotional predicates, which consist of verbs and adverbs. To demonstrate the feasibility of the ideas proposed in this paper, we performed a survey and compared the users’ responses with degrees of emotion estimated by the proposed method.     

A framework for context-aware self-adaptive mobile applications SPL

Mobile Applications are rapidly emerging as a convenient medium for using a variety of services. Over time and with the high penetration of smartphones in society, self-adaptation has become an essential capability required by mobile application users. In an ideal scenario, an application is required to adjust its behavior according to the current context of its use. This raises the challenge in mobile computing towards the design and development of applications that sense and react to contextual changes to provide a value-added user experience. In its general sense, context information can relate to the environment, the user, or the device status. In this paper, we propose a novel framework for building context aware and adaptive mobile applications. Based on feature modeling and Software Product Lines (SPL) concepts, this framework guides the modeling of adaptability at design time and supports context awareness and adaptability at runtime. In the core of the approach, is a feature meta-model that incorporates, in addition to SPL concepts, application feature priorities to drive the adaptability. A tool, based on that feature model, is presented to model the mobile application features and to derive the SPL members. A mobile framework, built on top of OSGI framework to dynamically adapt the application at runtime is also described.     

A pluggable middleware architecture for developing context-aware mobile applications

The proliferation of powerful smartphone devices provides a great opportunity for context-aware mobile applications becoming mainstream. However, we argue that conventional software development techniques suffer because of the added complexity required for collecting and managing context information. This paper presents a component-based middleware architecture which facilitates the development and deployment of context-aware applications via reusable components. The main contribution of this approach is the combination of a development methodology with the middleware architecture, which together bring significant value to developers of context-aware applications. Further contributions include the following: The methodology utilizes separation of concerns, thus decreasing the developmental cost and improving the productivity. The design and implementation of context-aware applications are also eased via the use of reusable components, called context plug-ins. Finally, the middleware architecture facilitates the deployment and management of the context plug-ins in a resource-aware manner. The proposed methodology and middleware architecture are evaluated both quantitatively and qualitatively.