User click prediction for personalized job recommendation

Major job search engines aggregate tens of millions of job postings online to enable job seekers to find valuable employment opportunities. Predicting the probability that a given user clicks on jobs is crucial to job search engines as the prediction can be used to provide personalized job recommendations for job seekers. This paper presents a real-world job recommender system in which job seekers subscribe to email alert to receive new job postings that match their specific interests. The architecture of the system is introduced with the focus on the recommendation and ranking component. Based on observations of click behaviors of a large number of users in a major job search engine, we develop a set of features that reflect the click behavior of individual job seekers. Furthermore, we observe that patterns of missing features may indicate various types of job seekers. We propose a probabilistic model to cluster users based on missing features and learn the corresponding prediction models for individual clusters. The parameters in this clustering-prediction process are jointly estimated by EM algorithm. We conduct experiments on a real-world testbed by comparing various models and features. The results demonstrate the effectiveness of our proposed personalized approach to user click prediction.

     

Pyramid collaborative filtering technique for an intelligent autonomous guide agent

This article presents an autonomous guide agent that can observe a community of learners on the web, interpret the learners' inputs, and then assess their sharing. The goal of this agent is to find a reliable helper (tutor or other learner) to assist a learner in solving his task. Despite the growing number of Internet users, the ability to find helpers is still a challenging and important problem. Although helpers could have much useful information about courses to be taught, many learners fail to understand their presentations. For that, the agent must be able to deal autonomously with the following challenges: Do helpers have information that the learners need? Will helpers present information that learners can understand? And can we guarantee that these helpers will collaborate effectively with learners? We have developed a new filtering framework, called a pyramid collaborative filtering model, to whittle the number of helpers down to just one. We have proposed four levels for the pyramid. Moving from one level to another depends on three filtering techniques: domain model filtering, user model filtering, and credibility model filtering. A new technique is filtering according to helpers' credibilities. Our experiments show that this method greatly improves filtering effectiveness. © 2007 Wiley Periodicals, Inc. Int J Int Syst 22: 1065–1082, 2007.     

The application of netnography to the online dating service experiences of female users

ABSTRACT

The primary purpose of online dating sites, a specialised form of social media, is to aid their users in finding love and making friends. However, while such sites are very popular, only limited research has been conducted to investigate the experiences of female dating site users. Moreover, the factors underlying the popularity of online dating sites have also yet to be fully clarified. Therefore, the current study utilised netnography and online interviews to explore the experiences of female dating site users, while also observing the physical travel paths of such users within the field. More specifically, the particular situational context of online dating was investigated through an approach that included the following three stages: the observation and collection of online data, participation in an active manner, and emergent design. The study’s key theoretical contribution is its establishment of a netnography model along with eight associated propositions. Overall, the study’s findings substantially illuminate the online dating site experiences of female site users.     

Design and evaluation of an adaptive incentive mechanism for sustained educational online communities

Most online communities, such as discussion forums, file-sharing communities, e-learning communities, and others, suffer from insufficient user participation in their initial phase of development. Therefore, it is important to provide incentives to encourage participation, until the community reaches a critical mass and “takes off”. However, too much participation, especially of low-quality can also be detrimental for the community, since it leads to information overload, which makes users leave the community. Therefore, to regulate the quality and the quantity of user contributions and ensure a sustainable level of user participation in the online community, it is important to adapt the rewards for particular forms of participation for individual users depending on their reputation and the current needs of the community. An incentive mechanism with these properties is proposed. The main idea is to measure and reward the desirable user activities and compute a user participation measure, then cluster the users based on their participation measure into different classes, which have different status in the community and enjoy special privileges. For each user, the reward for each type of activity is computed dynamically based on a model of community needs and an individual user model. The model of the community needs predicts what types of contributions (e.g. more new papers or more ratings) are most valuable at the current moment for the community. The individual model predicts the style of contributions of the user based on her past performance (whether the user tends to make high-quality contributions or not, whether she fairly rates the contributions of others). The adaptive rewards are displayed to the user at the beginning of each session and the user can decide what form of contribution to make considering the rewards that she will earn. The mechanism was evaluated in an online class resource-sharing system, Comtella. The results indicate that the mechanism successfully encourages stable and active user participation; it lowers the level of information overload and therefore enhances the sustainability of the community.     

A bi-objective model to optimize reliability and cost of system with a choice of redundancy strategies

Reliability problems are an important type of optimization problems that are motivated by different needs of real-world applications such as telecommunication systems, transformation systems, and electrical systems, so on. This paper studies a special type of these problems which is called redundancy allocation problem (RAP) and develops a bi-objective RAP (BORAP). The model includes non-repairable series–parallel systems in which the redundancy strategy is considered as a decision variable for individual subsystems. The objective functions of the model are (1) maximizing system reliability and (2) minimizing the system cost. Meanwhile, subject to system-level constraint, the best redundancy strategy among active or cold-standby, component type, and the redundancy level for each subsystem should be determined. To have a more practical model, we have also considered non-constant component hazard functions and imperfect switching of cold-standby redundant component. To solve the model, since RAP belong to the NP-hard class of the optimization problems, two effective multi-objective metaheuristic algorithms named non-dominated sorting genetic algorithms (NSGA-II) and multi-objective particle swarm optimization (MOPSO) are proposed. Finally, the performance of the algorithms is analyzed on a typical case and conclusions are demonstrated.     

Investigating critical success factors in online learning environments in higher education systems in the Caribbean*

ABSTRACT

Online learning has grown as a key method in education management over the last couple of decades. Studies have shown that significant investments in this technology are being made by universities, yet the full benefits expected have not been realized due to issues and challenges experienced by stakeholders such as learners and instructors in adopting and effectively using e-learning. This is especially true in developing economies where they may be attempting online delivery modes for the first time. In this study, we explore the question “What are the factors that influence university students’ adoption and use of an e-learning system in the context of the English-speaking Caribbean?” using an extended technology acceptance model framework. Partial least squares analysis was used to test the derived research model and found that critical success factors influencing students’ perception and use in online learning settings, particularly those within a developing economy conditions, include supportive cultural practices, access to computers, system or online environment availability, computer and online learning self-efficacy, user perception of usefulness and ease of use. These results have significant implications for university executives and policy-makers as they consider adopting online learning delivery modalities for users.     

Videolization: knowledge graph based automated video generation from web content

Web content nowadays can also be accessed through new generation of Internet connected TVs. However, these products failed to change users’ behavior when consuming online content. Users still prefer personal computers to access Web content. Certainly, most of the online content is still designed to be accessed by personal computers or mobile devices. In order to overcome the usability problem of Web content consumption on TVs, this paper presents a knowledge graph based video generation system that automatically converts textual Web content into videos using semantic Web and computer graphics based technologies. As a use case, Wikipedia articles are automatically converted into videos. The effectiveness of the proposed system is validated empirically via opinion surveys. Fifty percent of survey users indicated that they found generated videos enjoyable and 42 % of them indicated that they would like to use our system to consume Web content on their TVs.

     

Modelling and performance analysis of content sharing and distribution in community networks with infrastructure support

We consider the problem of sharing a viral file between users in a local community network (e.g., college and office campuses). Community computing is computer networking among and between users in a geographically bounded setting for local purposes and activities. Due to the community-oriented nature of such networks, it is likely that users of a community network would like to share content. Peer-to-Peer (P2P) networks have turned out to be one of the most innovative paradigms for sharing content on the Internet. In this paper, we analyze the performance of P2P content sharing in community networks and investigate the role that infrastructure nodes (helpers) can play to enhance the performance of content sharing and distribution. We model the evolution of content demand in a community network. The use of this demand prediction model allows us to design a delicate P2P-with-helpers content distribution system. Our insights that we obtain using fluid-flow model increase our understanding of how helper provisioning affects the performance of content sharing and distribution. The derived results show that significant reduction in both the cost of distributing content and the average content download time can be realized when only few infrastructure nodes in the community network play the role helpers and cache P2P objects.     

Designing Internet-Based Payment Systems: Guidelines and Empirical Basis

ABSTRACT

This article describes research into online electronic payment systems, focusing on the aspects of payment systems that are critical for their acceptance by end users. Based on our earlier research and a diary study of payments with an online payment system and with online banking systems of a reputable bank, we proposed a set of 12 interaction design guidelines. The guidelines have been applied during the implementation and redesign of a new payment system. An extensive experimental comparison of the original version of the system with the one designed according to the design guidelines has provided confirmation of the relevance and adequacy of these guidelines for designing online payment systems.     

Understanding the impact of m-banking on individual performance: DeLone & McLean and TTF perspective

Mobile banking (m-banking) is an expanding application of mobile commerce that has claimed the attention and interest of e-commerce researchers. One of the most welcome recent developments in m-banking has been the growing interest in end-user use, user satisfaction, and individual performance. We propose a model combining the DeLone & McLean IS success model and the Task Technology Fit (TTF) model to evaluate the impact of m-banking on individual performance. The empirical approach is based on an online survey questionnaire of 233 individuals. The results reveal that use and user satisfaction are important precedents of individual performance, and the importance of the moderating effects of TTF over usage to individual performance. The system quality, information quality, and service quality positively affect user satisfaction. Understanding the significance of m-banking context on individual performance is useful to provide new insight to m-banking managers to apply strategies to retain users or even attract potential adopters. We provide the theoretical and practical implications of our findings.     

Microblogs recommendations based on implicit similarity in content social networks

With the development of online social networking applications, microblogs have become a necessary online communication network in daily life. Users are interested in obtaining personalized recommendations related to their tastes and needs. In some microblog systems, tags are not available, or the use of tags is rare. In addition, user-specified social relations are extremely rare. Hence, sparsity is a problem in microblog systems. To address this problem, we propose a new framework called Pblog to alleviate sparsity. Pblog identifies users’ interests via their microblogs and social relations and computes implicit similarity among users using a new algorithm. The experimental results indicated that the use of this algorithm can improve the results. In online social networks, such as Twitter, the number of microblogs in the system is high, and it is constantly increasing. Therefore, providing personalized recommendations to target users requires considerable time. To address this problem, the Pblog framework groups similar users using the analytic hierarchy process (AHP) method. Then, Pblog prunes microblogs of the target user group and recommends microblogs with higher ratings to the target user. In the experimental results section, the Pblog framework was compared with several other frameworks. All of these frameworks were run on two datasets: Twitter and Tumblr. Based on the results of these comparisons, the Pblog framework provides more appropriate recommendations to the target user than previous frameworks.

     

Discerning individual interests and shared interests for social user profiling

Traditionally, research about social user profiling assumes that users share some similar interests with their followees. However, it lacks the studies on what topic and to what extent their interests are similar. Our study in online sharing sites reveals that besides shared interests between followers and followees, users do maintain some individual interests which differ from their followees. Thus, for better social user profiling we need to discern individual interests (capturing the uniqueness of users) and shared interests (capturing the commonality of neighboring users) of the users in the connected world. To achieve this, we extend the matrix factorization model by incorporating both individual and shared interests, and also learn the multi-faceted similarities unsupervisedly. The proposed method can be applied to many applications, such as rating prediction, item level social influence maximization and so on. Experimental results on real-world datasets show that our work can be applied to improve the performance of social rating. Also, it can reveal some interesting findings, such as who likes the “controversial” items most, and who is the most influential in attracting their followers to rate an item.     

基于社会选择理论的在线服务评价

用户评价标准不一致和偏好不一致导致网络空间中的在线服务之间不具备公正的可比较性,从而用户难以选择到满意的在线服务,因此,提出了基于社会选择理论计算在线服务优劣的排序方法。首先,根据用户给出的用户-服务评价矩阵构建群体偏好矩阵;然后,基于群体偏好矩阵和Kemeny社会选择函数构建0-1整数规划模型;最后,通过求解该模型可得到服务的最优排序结果。该方法聚合个体偏好为群体偏好,决策符合群体大多数人的偏好且与个体偏好保持最大的一致性。通过理论分析和实验验证了该方法的合理性和有效性。实验结果表明,该方法能有效地解决在线服务之间的不可比较性问题,实现在线服务的优劣排序,并可以有效抵制推荐攻击,具有较强的抗操纵性。     

A hybrid online-product recommendation system: Combining implicit rating-based collaborative filtering and sequential pattern analysis

Many online shopping malls in which explicit rating information is not available still have difficulty in providing recommendation services using collaborative filtering (CF) techniques for their users. Applying temporal purchase patterns derived from sequential pattern analysis (SPA) for recommendation services also often makes users unhappy with the inaccurate and biased results obtained by not considering individual preferences. The objective of this research is twofold. One is to derive implicit ratings so that CF can be applied to online transaction data even when no explicit rating information is available, and the other is to integrate CF and SPA for improving recommendation quality. Based on the results of several experiments that we conducted to compare the performance between ours and others, we contend that implicit rating can successfully replace explicit rating in CF and that the hybrid approach of CF and SPA is better than the individual ones.     

Recommender system with grey wolf optimizer and FCM

Recommender systems are contributing a significant aspect in information filtering and knowledge management systems. They provide explicit and reliable recommendations to the users so that user can get information about all products in e-commerce domain. In the era of big data and large complex information delivery system, it is impossible to get the right information in the online environment. In this research work, we offered a novel movie-based collaborative recommender system which utilizes the bio-inspired gray wolf optimizer algorithm and fuzzy c-mean (FCM) clustering technique and predicts rating of a movie for a particular user based on his historical data and similarity of users. Gray wolf optimizer algorithm was applied on the Movielens dataset to obtain the initial clusters, and also the initial positions of clusters are obtained. FCM is used to classify the users in the dataset by similarity of user ratings. Our proposed collaborative recommender system performed extremely well with respect to accuracy and precision. We analyzed our proposed recommender system over Movielens dataset which is available publically. Various evaluation metrics were utilized such as mean absolute error, standard deviation, precision and recall. We also compared the performance of projected system with already established systems. The experiment results delivered by proposed recommender system demonstrated that efficiency and performance are enhanced and also offered better recommendations when compared with our previous work [1].

     

Experiments Toward a Mutual Adaptive Speech Interface That Adopts the Cognitive Features Humans Use for Communication and Induces and Exploits Users' Adaptations

Interactive agents such as pet robots or adaptive speech interface systems that require forming a mutual adaptation process with users should have two competences. One of these is recognizing reward information from users' expressed paralanguage information, and the other is informing the learning system about the users by means of that reward information. The purpose of this study was to clarify the specific contents of reward information and the actual mechanism of a learning system by observing how 2 persons could create a smooth speech communication, such as that between owners and their pets.

A communication experiment was conducted to observe how human participants create smooth communication through acquiring meaning from utterances in languages they did not understand. Then, based on experimental results, a meaning-acquisition model that considers the following 2 assumptions was constructed: (a) To achieve a mutual adaptive relationship with users, the model needs to induce users' adaptation and to exploit this induced adaptation to recognize the meanings of a user's speech sounds; and (b) to recognize users' utterances through trial-and-error interaction regardless of the language used, the model should focus on prosodic information in speech sounds, rather than on the phoneme information on which most past interface studies have focused.

The results confirmed that the proposed model could recognize the meanings of users' verbal commands by using participants' adaptations to the model for its meaning-acquisition process. However, this phenomenon was observed only when an experimenter gave the participants appropriate instructions equivalent to catchphrases that helped users learn how to use and interact intuitively with the model. Thus, this suggested the need for a subsequent study to discover how to induce the participants' adaptations or natural behaviors without giving these kinds of instructions.     

Emotional support and suicide prevention through the Internet: A field project report

The Internet can be efficiently exploited to help people in severe emotional distress, including those contemplating suicide. Based on this premise and guided by various psychological principles that characterize Internet communication and behavior, on the one hand, and by factors related to the provision of emotional support, on the other, an Israeli project with the Hebrew acronym SAHAR was established. The idea behind SAHAR was to initiate an anonymous, confidential online environment that would attract people in a crisis situation and offer them a listening ear, mental support, and warmth, provided by anonymous, skilled helpers. At the heart of this exclusively online service is a content-rich Hebrew website (http://www.sahar.org.il) that provides relevant and continuously updated information for people in need. Accessed through the site, SAHAR offers, in addition, personal communication to users through synchronous and asynchronous support. For group communication, SAHAR provides online forums (a chat room will soon be launched). The website is accessed more than 10,000 times a month, or 350 times a day, a considerable number relative to Israel’s small population. Approximately 1000 personal contacts with SAHAR itself take place, each month, of which at least a third of the distressed users are clearly suicidal. The forums receive over 200 new messages a day. SAHAR on numerous occasions has participated in rescue operations of individuals who threatened to commit suicide or were actually in the process of trying. In many other cases, a supportive conversation or a referral to appropriate help resources prevented hasty decisions by highly distressed, desperate people contemplating suicide. Feedback by users also indicates the success of SAHAR as a unique psychological application on the Internet.     

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.

     

The effects of an active development of the mental model in the training process: experimental results in a word processing system

Abstract

Three different training programmes for a word processing system were experimentally compared: (1) a sequential programme, which taught low-level skills and which did not help the user actively to develop a mental model, (2) a hierarchical programme, which provided an explicit and integrated conceptual model of the system to the user and (3) a programme in which the users were asked to develop hypotheses on the functioning of the software and in which they were encouraged to use an active and exploratory approach. From an action theory point of view it was hypothesized that the third group would be superior to the first group. In an experimental study with two training sessions each of two hours and a two-hour testing session (n = 15), this was shown to be the case for several performance criteria (error time, transfer and experimenter rating). Additionally, an interindividual difference variable lo measure the individual learning style was used, giving results that could be interpreted in a similar way to the experimental results.