Individual User Preferences Research Articles

Semantic Approach for Multifarious Ranked Novelty and Diversity in News Recommendations

Information overload on web creates lots of inconvenience for end users to unearth requisite information which instigate the demand of personalized recommendations. Recommender systems strive to attain just accuracy in recommendations based on the history of user preferences, resulting in over specialized recommendations. This leads to gradual loss of user’s interest in the system. These approaches also fail to recommend other products in long list which could be of user interest but user is not aware of. This in-turn leads to sparse user profiles.In dynamic domains, novelty and diversity in coverage can handle over specialization problem. To bring useful novelty and diversity, it is proposed to semantically expand the user preferences. For this purpose our designed multi level ontology is used with linkages among concepts, entities and properties, carrying different weights. Ontology is also enriched with different relations from online semantic lexicon and annotated with additional information about entities from open and linked external knowledge source.For relevant accurate recommendations in highly dynamic domains, we propose semantic approach to implicitly capture individual user preferences in separate profiles. One profile is to capture short term dynamic interests with temporal effects and another is to capture long term static interests. Semantic profiling helps to handle preferences in manageable number of news categories which are arranged hierarchically. It also helps to handle synonymy and polysemy problems, reducing ambiguity in profiling. An additional important step of outlier analysis and rectification is proposed to handle the effects of sudden unexpected temporary drifts in dynamic user preferences. This focused analysis helps to maintain correct ranking of user preferences which is generally overlooked. The rectified ranking of user interests is fetched for semantic expansion of user preferences, to bring correct ranking in diversity and novelty.Proposed approach semantically brings reasonable multifarious novelty and diversity by analyzing different features of item. Testing is based on live data gathered from RSS feeds of popular news providers for trustworthiness. Transparency is achieved by presenting new recommendations in separate sections along with explicit trace path options, based on difference in approaches. Transparency assists end user in making decision. User’s interests in diverse and novel recommendations are updated in their individual profiles, resulting in reduction of sparsity.

Let’s Go Out For A Day Trip? Perspectives of Psychedelics (Ab)Users on the Safety of Acid (LSD) Tripping in Public Places

BACKGROUND: Novel psychoactive substances (NPS) represent a unique phenomenon of the 21st century. These substances are of critical consequences on public health and national economies. Hallucinogens, also known as psychedelics and entheogens, represent one category of NPS. Numerous private groups do exist on the online drug fora and the online social platforms including Facebook. Psychedelic tripping or acid trip (using LSD) depicts one of the controversial life experiences; tripping can be indoor or in public.MATERIALS And METHODS: This study is observational and cross-sectional; it was based on an Internet Snapshot taken for a private group on Facebook; the group is dedicated for (ab)users of psychedelic substances. The snapshot was captured for a thread in relation to a critical question which was posted on the safety of day tripping using acid (LSD) in public places. Individual accounts of commenters (n=172) were analyzed in relation to demographics, length and themes of comments, and the attitude towards public tripping. This study is the first of its kind; it aims to conclude with an inference whether outdoor tripping is favourable by psychedelics users or not.RESULTS: A total of 137 psychedelic users’ comments were analyzed out of 172 (79.7%); males contributed more (n=111, 81%); the mean age was 32.14 years; most were Caucasian males from the US. (Ab)users were mainly geo-mapped into the US (85.4%), Canada (5.1%), and UK (3.7%). Those who had a positive attitude in relation to day tripping public places accounted for three-quarters (75.2%). Ethnicities and nationalities had no differential effect on a psychedelic user’s age nor his (her) enthusiasm for day tripping. However, (ab)users from the US were found to be more enthused. Further, there was a significant difference in relation to the attitude in between individuals with; positive attitude and negative attitude (p-value<0.001).CONCLUSION: Psychedelics (ab)users appeared to be in favour of having an acid trip in public; those were mainly geographically mapped into the developed countries, while the contribution of the developing countries was minimal. This study can be copied to populations of interest of different backgrounds, cultures, and ethnicities in an aim to infer changes in trends and preferences of individual users across time and place.

Lighting preferences in individual offices

Abstract Workplaces with good daylighting offer visual comfort to users, give them a series of physiological and psychological benefits and allow good performance of visual activities, besides saving energy. However, this solution is not always adopted: lighting type preferences involve many variables besides the availability of daylight. This paper explores a case study through the analysis of questionnaire answers and computer simulations of a series of metrics related to quality of lighting with the aim of finding explanations for the lighting preferences of individual office users. The results show that, although the offices present good daylighting conditions and no glare potential, and users are satisfied with daylighting, these parameters are not sufficient to explain the predominant lighting preferences. The findings have also shown that there is no consensus about which parameters potentially cause visual comfort, while the parameters that cause discomfort are clearly identified. In addition, in this study, 49% of the preference for mixed lighting (daylight plus electrical light) can be explained by the fact that mixed lighting produces better modeling than daylighting alone.

Sensor-driven, human-in-the-loop lighting control

Smart indoor lighting systems use occupancy and light sensor data to adapt artificial lighting in accordance with changing occupancy and daylight conditions. Such systems can be designed to reduce lighting energy consumption significantly. However, these systems cannot account for individual user preferences at the workplace in real time. We propose a sensor-driven, human-in-the-loop lighting system that incorporates user feedback in addition to occupancy and light sensor inputs. In this system, luminaires transmit unique visible light communication identifier signals. By processing the image captured by a smartphone camera, a user obtains two pieces of information: visible light communication identifiers of luminaires in the vicinity and average image pixel value. A control algorithm is designed that incorporates these user inputs along with occupancy and light sensor inputs to determine the dimming levels of the luminaires to achieve illumination levels acceptable to users. We compare the performance of the proposed lighting control system with a sensor-driven lighting control system in an office test bed.

A Hybrid Collaborative Filtering Model with Deep Structure for Recommender Systems

Collaborative filtering (CF) is a widely used approach in recommender systems to solve many real-world problems. Traditional CF-based methods employ the user-item matrix which encodes the individual preferences of users for items for learning to make recommendation. In real applications, the rating matrix is usually very sparse, causing CF-based methods to degrade significantly in recommendation performance. In this case, some improved CF methods utilize the increasing amount of side information to address the data sparsity problem as well as the cold start problem. However, the learned latent factors may not be effective due to the sparse nature of the user-item matrix and the side information. To address this problem, we utilize advances of learning effective representations in deep learning, and propose a hybrid model which jointly performs deep users and items’ latent factors learning from side information and collaborative filtering from the rating matrix. Extensive experimental results on three real-world datasets show that our hybrid model outperforms other methods in effectively utilizing side information and achieves performance improvement.

Personalized Group Recommender Systems for Location- and Event-Based Social Networks

Location-Based Social Networks (LBSNs) such as Foursquare, Google+ Local, and so on, and Event-Based Social Networks (EBSNs) such as Meetup, Plancast, and so on, have become popular platforms for users to plan, organize, and attend social events with friends and acquaintances. These LBSNs and EBSNs provide rich content such as online and offline user interactions, location/event descriptions that can be leveraged for personalized group recommendations. In this article, we propose novel Collaborative Filtering-based Bayesian models to capture the location or event semantics and group dynamics such as user interactions, user group membership, user influence, and the like for personalized group recommendations. Empirical experiments on two large real-world datasets (Gowalla LBSN dataset and Meetup EBSN dataset) show that our models outperform the state-of-the-art group recommender systems. We discuss the group characteristics of our datasets and show that modeling of group dynamics learns better group preferences than aggregating individual user preferences. Moreover, our model provides human interpretable results that can be used to understand group participation behavior and location/event popularity.

A Trainable Hearing Aid Algorithm Reflecting Individual Preferences for Degree of Noise-Suppression, Input Sound Level, and Listening Situation.

Objectives In an effort to improve hearing aid users’ satisfaction, recent studies on trainable hearing aids have attempted to implement one or two environmental factors into training. However, it would be more beneficial to train the device based on the owner’s personal preferences in a more expanded environmental acoustic conditions. Our study aimed at developing a trainable hearing aid algorithm that can reflect the user’s individual preferences in a more extensive environmental acoustic conditions (ambient sound level, listening situation, and degree of noise suppression) and evaluated the perceptual benefit of the proposed algorithm.Methods Ten normal hearing subjects participated in this study. Each subjects trained the algorithm to their personal preference and the trained data was used to record test sounds in three different settings to be utilized to evaluate the perceptual benefit of the proposed algorithm by performing the Comparison Mean Opinion Score test.Results Statistical analysis revealed that of the 10 subjects, four showed significant differences in amplification constant settings between the noise-only and speech-in-noise situation (P<0.05) and one subject also showed significant difference between the speech-only and speech-in-noise situation (P<0.05). Additionally, every subject preferred different β settings for beamforming in all different input sound levels.Conclusion The positive findings from this study suggested that the proposed algorithm has potential to improve hearing aid users’ personal satisfaction under various ambient situations.

사물 인터넷 환경에서의 그룹 사용자를 위한 그룹 구성 정보 기반 서비스 추천 방법

Recommender systems can be used to assist users in selecting required services for their tasks in Internet of Things (IoT) environments in which diverse services can be provided by utilizing IoT devices. Traditional research on recommendation mainly focuses on predicting preferences of individual users. However, in IoT environments, not only individual users but also groups of users can access services in the environments. In this study, we analyzed user groups' preferences on services and developed service recommendation approach for new groups that do not have a history of accessing IoT-services in a certain place. Our approach extends the traditional user-based collaborative filtering by considering the similarity between user groups based on their member organization. We conducted experiments with a real-world dataset collected from IoT testbed environments. The results demonstrate that the proposed approach is effective to recommend services to new user groups in IoT environments.

Person-Centered Multimedia Computing: A New Paradigm Inspired by Assistive and Rehabilitative Applications

Human-centered multimedia computing (HCMC) focuses on a tight engagement of humans in the design, development, and deployment of multimedia solutions. However, people's abilities change over time due to a variety of reasons, including age, context, and geographical location. To address this challenge, the authors recently introduced the concept of person-centered multimedia computing, where the emphasis is on understanding the individual user's preferences and expectations toward designing, developing, and deploying effective solutions. Today's multimedia technology is largely geared toward the able population; individuals with disabilities have largely been absent in the design process and thus must adapt themselves (often unsuccessfully) to available solutions. Further, individuals with disabilities have specific and individualized requirements that necessitate a person-centered, adaptive approach to multimedia computing. Here, the authors present the person-centered multimedia computing approach inspired by assistive and rehabilitative applications.

A web-based multicriteria evaluation of spatial trade-offs between environmental and economic implications from hydraulic fracturing in a shale gas region in Ohio

Planning of shale gas infrastructure and drilling sites for hydraulic fracturing has important spatial implications. The evaluation of conflicting and competing objectives requires an explicit consideration of multiple criteria as they have important environmental and economic implications. This study presents a web-based multicriteria spatial decision support system (SDSS) prototype with a flexible and user-friendly interface that could provide educational or decision-making capabilities with respect to hydraulic fracturing site selection in eastern Ohio. One of the main features of this SDSS is to emphasize potential trade-offs between important factors of environmental and economic ramifications from hydraulic fracturing activities using a weighted linear combination (WLC) method. In the prototype, the GIS-enabled analytical components allow spontaneous visualization of available alternatives on maps which provide value-added features for decision support processes and derivation of final decision maps. The SDSS prototype also facilitates nonexpert participation capabilities using a mapping module, decision-making tool, group decision module, and social media sharing tools. The logical flow of successively presented forms and standardized criteria maps is used to generate visualization of trade-off scenarios and alternative solutions tailored to individual user's preferences that are graphed for subsequent decision-making.

A Context-awareness Based Dynamic Personalized Hierarchical Ontology Modeling Approach

This paper proposes a new approach to construct an efficient and dynamic updated ontology model. In order to capture the implicit context information and to implement the personalized scalability demands of context model for different users, this paper proposes a new approach to construct a context-awareness based dynamic personalized hierarchical ontology model which divides the ontology into 2 layers: the first layer (general context-aware ontology) is used to capture the most essential conceptual entities in the context-aware computing environment, and the second layer (personalized ontology) is used to capture the individual user's preference by considering the domain ontology, knowledge structure, the term co-occurrence frequency and the Active Degree of each device. An algorithm of Domain Classification Algorithm is used and an Improved Concept Identification Algorithm is proposed in this research. The simulation results show that the time overhead of the hierarchical ontology model is low and it provides the users with results that more accurately satisfy their specific goal and intent.

인터넷 쇼핑몰 니즈 분석 시스템의 설계 및 구현

ABSTRACT Even though users choose goods they want to buy in on-line shopping malls, real purchase is often performed in off-line shopping malls. It is called reverse showrooming. It means that users’ analysis of goods based on images and description of internet shopping malls has limitation. Thus, large-scale online shopping malls provide a customized shopping information. However, in that case, the provided information is a simple list of goods users bought or retrieved. Thus, a system to analyze various needs of users and apply the result into on-line shopping mall is necessary. In this paper, an analysis system is proposed. The system contains a module to analyze user defined preference and a module to analyze users’ reviews. The former designates two goods and collects preferences of individual users. the latter analyzes reviews about purchased goods based on database dictionary stored in advance for analyzing reviews. The system implemented shows that it is possible to recommend some goods that meet each users’s needs

A Low Cost, Highly Scalable Wireless Sensor Network Solution to Achieve Smart LED Light Control for Green Buildings

Reducing energy demand in the residential and industrial sectors is an important challenge worldwide. In particular, lights account for a great portion of total energy consumption, and unfortunately a huge amount of this energy is wasted. Light-emitting diode (LED) lights are being used to light offices, houses, industrial, or agricultural facilities more efficiently than traditional lights. Moreover, the light control systems are introduced to current markets, because the installed lighting systems are outdated and energy inefficient. However, due to high costs, installation issues, and difficulty of maintenance; existing light control systems are not successfully applied to home, office, and industrial buildings. This paper proposes a low cost, wireless, easy to install, adaptable, and smart LED lighting system to automatically adjust the light intensity to save energy and maintaining user satisfaction. The system combines motion sensors and light sensors in a low-power wireless solution using Zigbee communication. This paper presents the design and implementation of the proposed system in a real-world deployment. Characterization of a commercial LED panel was performed to evaluate the benefit of dimming for this light technology. Measurements of total power consumption over a continuous six months period (winter to summer) of a busy office were acquired to verify the performance and the power savings across several weather conditions scenarios. The proposed smart lighting system reduces total power consumption in the application scenario by 55% during a six month period and up to 69% in spring months. These figures take also into account individual user preferences.

Mobile Information Recommendation Using Multi-Criteria Decision Making with Bayesian Network

The advancement of network technology and the popularization of the Internet lead to increased interest in information recommendation. This paper proposes a group recommendation system that takes the preferences of group users in mobile environment and applies the system to recommendation of restaurants. The proposed system recommends the restaurants by considering various preferences of multiple users. To cope with the uncertainty in mobile environment, we exploit Bayesian network, which provides reliable performance and models individual user's preference. Also, Analytical Hierarchy Process of multi-criteria decision-making method is used to estimate the group users' preference from individual users' preferences. Experiments in 10 different situations provide a comparison of the proposed method with random recommendation, simple rule-based recommendation and neural network recommendation, and confirm that the proposed method is useful with the subjective test.

Contextual preference mining for user profile construction

The emerging of ubiquitous computing technologies in recent years has given rise to a new field of research consisting in incorporating context-aware preference querying facilities in database systems. One important step in this setting is the Preference Elicitation task which consists in providing the user ways to inform his/her choice on pairs of objects with a minimal effort. In this paper we propose an automatic preference elicitation method based on mining techniques. The method consists in extracting a user profile from a set of user preference samples. In our setting, a profile is specified by a set of contextual preference rules verifying properties of soundness and conciseness. After proving that the problem is NP-complete, we propose a resolution in 2 phases. The first phase extracts all individual user preferences by means of contextual preference rules. The second phase builds the user profile starting from this collection of rules using a greedy method. To assess the quality of user profiles, we propose three ranking techniques benefiting from these profiles that enable us to rank objects according to user preferences. We evaluate the efficacy of our three ranking strategies and compare them with a well-known ranking method (SVMRank). The evaluation is carried out through an extensive set of experiments executed on a real-world database of user preferences about movies.

Online Estimation of Evolving Human Visual Interest

Regions in video streams attracting human interest contribute significantly to human understanding of the video. Being able to predict salient and informative Regions of Interest (ROIs) through a sequence of eye movements is a challenging problem. Applications such as content-aware retargeting of videos to different aspect ratios while preserving informative regions and smart insertion of dialog (closed-caption text) 1 into the video stream can significantly be improved using the predicted ROIs. We propose an interactive human-in-the-loop framework to model eye movements and predict visual saliency into yet-unseen frames. Eye tracking and video content are used to model visual attention in a manner that accounts for important eye-gaze characteristics such as temporal discontinuities due to sudden eye movements, noise, and behavioral artifacts. A novel statistical- and algorithm-based method gaze buffering is proposed for eye-gaze analysis and its fusion with content-based features. Our robust saliency prediction is instantiated for two challenging and exciting applications. The first application alters video aspect ratios on-the-fly using content-aware video retargeting, thus making them suitable for a variety of display sizes. The second application dynamically localizes active speakers and places dialog captions on-the-fly in the video stream. Our method ensures that dialogs are faithful to active speaker locations and do not interfere with salient content in the video stream. Our framework naturally accommodates personalisation of the application to suit biases and preferences of individual users.

Reverse k-ranks query

Finding matching customers for a given product based on individual user's preference is critical for many applications, especially in e-commerce. Recently, the reverse top- k query is proposed to return a number of customers who regard a given product as one of the k most favorite products based on a linear model. Although a few "hot" products can be returned to some customers via reverse top- k query, a large proportion of products (over 90%, as our example illustrates, see Figure 2) cannot find any matching customers. Inspired by this observation, we propose a new kind of query (R- k Ranks) which finds for a given product, the top- k customers whose rank for the product is highest among all customers, to ensure 100% coverage for any given product, no matter it is hot or niche . Not limited to e-commerce, the concept of customer - product can be extended to a wider range of applications, such as dating and job-hunting. Unfortunately, existing approaches for reverse top- k query cannot be used to handle R- k Ranks conveniently due to infeasibility of getting enough elements for the query result. Hence, we propose three novel approaches to efficiently process R- k Ranks query, including one tree-based method and two batch-pruning-based methods. Analysis of theoretical and experimental results on real and synthetic data sets illustrates the efficacy of the proposed methods.

Collaborative Filtering beyond the User-Item Matrix

Over the past two decades, a large amount of research effort has been devoted to developing algorithms that generate recommendations. The resulting research progress has established the importance of the user-item (U-I) matrix, which encodes the individual preferences of users for items in a collection, for recommender systems. The U-I matrix provides the basis for collaborative filtering (CF) techniques, the dominant framework for recommender systems. Currently, new recommendation scenarios are emerging that offer promising new information that goes beyond the U-I matrix. This information can be divided into two categories related to its source: rich side information concerning users and items, and interaction information associated with the interplay of users and items. In this survey, we summarize and analyze recommendation scenarios involving information sources and the CF algorithms that have been recently developed to address them. We provide a comprehensive introduction to a large body of research, more than 200 key references, with the aim of supporting the further development of recommender systems exploiting information beyond the U-I matrix. On the basis of this material, we identify and discuss what we see as the central challenges lying ahead for recommender system technology, both in terms of extensions of existing techniques as well as of the integration of techniques and technologies drawn from other research areas.

Stochastic optimization on social networks with application to service pricing

In this paper we develop a combined simulation and optimization approach for solving difficult decision problems on complex dynamic networks. For a specific reference problem we consider a telecommunication service provider who offers a telecommunication service to a market with network effects. More particularly, the service consumption of an individual user depends on both idiosyncratic characteristics and the popularity of this service among the customer’s immediate neighborhood. Both the social network and the individual user preferences are largely heterogeneous and changing over time. In addition the service provider’s decisions are made in absence of perfect knowledge about user preferences. The service provider pursues the strategy of stimulating the demand by offering differentiated prices to the customers. For finding the optimal pricing we apply a stochastic quasi-gradient algorithm that is integrated with a simulation model that drives the evolution of the network and user preferences over time. We show that exploiting the social network structure and implementing differentiated pricing can substantially increase the revenues of a service provider operating on a social network. More generally, we show that stochastic gradient methods represent a powerful methodology for the optimization of decisions in social networks.

Accessibility Support for Older Adults with the ACCESS Framework

Equitable access to the digital economy is predicated on the usability of the devices that are used to access electronic goods and services, with computers being the primary mechanism for many users by which this is currently done. For novice users with special interaction requirements, current arrangements for enabling accessibility support are suboptimal. Older users in particular require special consideration with regards to the design of software support packages to ensure the burden of knowledge required to configure a system is reasonable. This article describes the ACCESS Framework, a novel open-source, plug-in enabled software framework designed to address some of the issues around providing accessibility support on the desktop. The framework employs a system through which corrections are successively adapted to an individual user's preferences. Through empirical work with older adults, the framework has been shown to provide an understandable, appropriate, and effective way to enable accessibility support.