As repetition of activities can establish habits and exploration of new ones can provide a healthy variety, we investigate how a recommender system for physical activities can optimally balance these two approaches. We conducted an eight-week user study with 62 physically inactive participants who receive personalized repetition and exploration recommendations in a random order. We distinguish between location, workout, and general activities, and collect participants’ subjective perceptions. Our findings indicate that participants initially preferred exploring general activities, but rated repeating recommendations higher after two weeks. By exploring the optimal transition point from exploration to repetition in personalized recommendations, this study contributes to designing more effective recommender systems for health improvement and healthy habit formation.

As personalization has great potential to improve mobile health apps, analyzing the effect of different recommender algorithms in the health domain is still in its infancy. As such, this paper investigates whether more accurate recommendations from a content-based recommender or more diverse recommendations from a user-based collaborative filtering recommender will lead to more motivation to move. An eight-week longitudinal between-subject user study is being conducted with an Android app in which participants receive personalized recommendations for physical activities and tips to reduce sedentary behavior. The objective manipulation check confirmed that the group with collaborative filtering received significantly more diverse recommendations. The subjective manipulation check showed that the content-based group assigned more positive feedback for perceived accuracy and star rating to the recommendations they chose and executed. However, perceived diversity and inspiringness was significantly higher in the content-based group, suggesting that users might experience the recommendations differently. Lastly, momentary motivation for the executed activities and tips was significantly higher in the content-based group. As such, the preliminary results of this longitudinal study suggest that more accurate and less diverse recommendations have better effects on motivating users to move more.

Facing the climate change and the energy crisis, plenty of challenges remain to achieving the carbon neutrality in the energy intensive industry. This work investigates a flow-shop production scheduling problem minimizing the energy cost without delaying any jobs. The research problem comes from a Belgium printing company with solar panels as its own energy source. Two sub-problems, the job sequence determination and the job-machine allocation, are tackled using a memetic algorithm. An energy-efficient local search heuristic is also designed to make best use of the self-generated electricity. Validated using realistic production data of 12 jobs in a 2 day planning horizon, the proposed method outperforms a standard binary-encoded genetic algorithm in both solution quality (having 27.17% lower energy cost) and the calculation time (saving 69.89%).

This paper proposes a genetic algorithm (GA) combined with ray tracer to generate a cell-free topology of massive MIMO (mMIMO) for the optimal focusing performance serving multiple users. The realistic hardware impairment, for instance the non-ideal power amplifier, is taken into account of the system modeling and topology optimization. To the best of our knowledge, this is the first attempt to apply GA in optimizing the hardware-impaired multi-user cell-free mMIMO. Although the demonstrated numerical analysis is for indoor scenario, the proposed approach is transferable for generic scenarios. In GA, the base station (BS) antennas’ placement is encoded with an adjusted binary matrix representation, which is straightforward for the subsequent genetic operations. The explored candidates by GA can evolve beyond the parents, where the fitness of individuals is evaluated dynamically via a ray tracer radio channel simulator. Compared to the traditional GA, our proposed GA can find better solutions with a faster convergence speed. The algorithm provides near-optimal results in experiments, applicable to generic environment with multiple mobile users and different signal-to-interference-plus-noise ratios.

Continuous manufacturing is playing an increasingly important role in modern industry, while research on production scheduling mainly focuses on traditional batch processing scenarios. This paper provides an efficient genetic method to minimize energy cost, failure cost, conversion cost and tardiness cost involved in the continuous manufacturing. With the help of Industrial Internet of Things, a multi-objective optimization model is built based on acquired production and environment data. Compared with a conventional genetic algorithm, non-random initialization and elitist selection were applied in the proposed algorithm for better convergence speed. Problem specific constraints such as due date and precedence are evaluated in each generation. This method was demonstrated in the plant of a pasta manufacturer. In experiments of 71 jobs in a one-month window, near-optimal schedules were found with significant reductions in costs in comparison to the existing original schedule.

Price-based demand response stimulates factories to adapt their power consumption patterns to time-sensitive electricity prices, so that a rise in energy cost is prevented without affecting production on the shop floor. This paper introduces a multiobjective optimization (MOO) model that jointly schedules job processing, machine idle modes, and human workers under real-time electricity pricing. Beyond existing models, labor is considered due to a common tradeoff between energy cost and labor cost. An adaptive multiobjective memetic algorithm (AMOMA) is proposed to fast converge toward the Pareto front without loss in diversity. It leverages feedback of cross-dominance and stagnation in a search and a prioritized grouping strategy. In this way, adaptive balance remains between exploration of the nondominated sorting genetic algorithm II and exploitation of two mutually complementary local search operators. A case study of an extrusion blow molding process in a plastic bottle manufacturer and benchmarks demonstrate the MOO effectiveness and efficiency of AMOMA. The impacts of production-prohibited periods and relative portion of energy and labor costs on MOO are further analyzed, respectively. The generalization of this method was further demonstrated in a multimachine experiment. The common tradeoff relations between the energy and labor costs as well as between the makespan and the sum of the two cost parts were quantitatively revealed.

Nowadays, the energy price is rising. The consciousness of environmental sustainability of governments and customers has been ever increasing. Consequently, manufacturing enterprises are increasingly motivated to reduce the energy cost involved in their production activities. This paper proposes a novel production scheduling method to minimize the energy cost involved in the production at the unit process level. Compared to the emerging energy-conscious production scheduling methods, this method builds the finite state machine based energy model from power data that are measured from the shop floor. By following the formulated mixed integer linear programming model, the power states and changeovers of a unit process can be additionally scheduled, and the potential multiple process idle modes can be optimally selected between two jobs. In addition, the process power consumption behavior can be predicted along with the optimal schedule. This method was demonstrated in an extrusion blow molding process in a Belgian plastic bottle manufacturer. Compared to two conventional schedules, i.e., “as-early-as-possible” and “as-late-as-possible”, the schedule given by the proposed method is able to reduce 21% and 11% of electricity cost for completing the same production task before a due date.

Recommender systems have proven their usefulness as a tool to cope with the information overload problem for many online services offering movies, books, or music.Recommender systems rely on identifying individual users and deducing their preferences from the feedback they provide on the content.To automate this user identification and feedback process for TV applications, we propose a solution based on face detection and recognition services.These services output useful information such as an estimation of the age, the gender, and the mood of the person.Demographic characteristics (age and gender) are used to classify the user and cope with the cold start problem.Detected smiles and emotions are used as an automatic feedback mechanism during content consumption.Accurate results are obtained in case of a frontal view of the face.Head poses deviating from a frontal view and suboptimal illumination conditions may hinder face detection and recognition, especially if parts of the face, such as eyes or mouth are not sufficiently visible.

Rating datasets are of paramount importance in recommender systems research. They serve as input for recommendation algorithms, as simulation data, or for evaluation purposes. In the past, public accessible rating datasets were not abundantly available, leaving researchers no choice but to work with old and static datasets like MovieLens and Netflix. More recently, however, emerging trends as social media and smartphones are found to provide rich data sources which can be turned into valuable research datasets. While dataset availability is growing, a structured way for introducing and comparing new datasets is currently still lacking. In this work, we propose a five-step framework to introduce and benchmark new datasets in the recommender systems domain. We illustrate our framework on a new movie rating dataset—called MovieTweetings—collected from Twitter. Following our framework, we detail the origin of the dataset, provide basic descriptive statistics, investigate external validity, report the results of a number of reproducible benchmarks, and conclude by discussing some interesting advantages and appropriate research use cases.

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.However, in many applications, this contextual information may be relevant for the human decision process, and as a result, be important to incorporate into the recommendation process, which gave rise to context-aware recommender systems.However, the information value of contextual data for the recommendation process is highly dependent on the application domain and the users' consumption behavior in different contextual situations.This research aims to assess the information value of context for a recommender system of a mobile news service by analyzing user interactions and feedback.A large-scale user study shows that context-aware recommendations outperform traditional recommendations, but also indicates that the accuracy improvement might be limited in a real-life situation.Service usage takes place in a limited number of different contexts due to user habits and repetitive behavior, leaving little room for optimization based on the context.Data fragmentation over different contextual situations strengthens the sparsity problem, thereby limiting the user-perceived accuracy gain obtained by incorporating context in the recommender.These findings are important for news providers when considering to offer context-aware recommendations.

Mobile Shopping Applications (MSAs) are rapidly gaining popularity. They enhance the shopping experience, by offering customized recommendations or incorporating customer loyalty programs. Although MSAs are quite effective at attracting new customers and binding existing ones to a retailer's services, existing MSAs have several shortcomings. The data collection practices involved in MSAs and the lack of transparency thereof are important concerns for many customers. This paper presents inShopnito, a privacy-preserving mobile shopping application. All transactions made in inShopnito are unlinkable and anonymous. However, the system still offers the expected features from a modern MSA. Customers can take part in loyalty programs and earn or spend loyalty points and electronic vouchers. Furthermore, the MSA can suggest personalized recommendations even though the retailer cannot construct rich customer profiles. These profiles are managed on the smartphone and can be partially disclosed in order to get better, customized recommendations. Finally, we present an implementation called inShopnito, of which the security and performance is analyzed. In doing so, we show that it is possible to have a privacy-preserving MSA without having to sacrifice practicality.

In this research, a food recommendation strategy for patients in a care facility is proposed. Since many of these patients cannot express their personal preferences, a recommender system can assist the caregivers in the selection of the menu items that match the patients' preferences. Recommendations are generated based on three information sources: explicit ratings for menu items, implicit feedback based on the patient's eating behavior and the amount of food that was eaten, and inferred preferences for the ingredients of the menu items. In addition, monitoring the amount of food that was eaten by each patient can provide insights into the optimal amount of each menu item that has to be served to each patient. Furthermore, monitoring food consumption allows to detect irregularities in the eating behavior of the patient, which can be a symptom of illness.

Though most recommender systems make suggestions for individual users, in many circumstances the selected items (e.g., movies) are not for personal usage but rather for consumption in group. In this paper, we present a recommender system for audio-visual content that generates suggestions for groups of people (such as families or friends) in the home environment. In this context, different group recommendation strategies are evaluated for various algorithms and sizes of the group. An offline evaluation proves the assumption that for randomly composed groups the accuracy of all recommendation algorithms decreases if the group size grows. Besides, the results show that the group recommendation strategy which produces the most accurate results is depending on the algorithm that is used for generating individual recommendations. Consequently, if an existing recommender system for individuals is extended to a recommender system for groups, the group recommendation strategy has to be chosen based on the utilized recommendation algorithm in order to maximize the efficiency of the group recommendations.

The efficiency of personal suggestions generated by collaborative filtering techniques is highly dependent on the quality and quantity of the available consumption data. Extending data sets with additional consumption data (from the past) might enrich the user profiles and generally leads to more accurate recommendations. Although if a considerable amount of profile information is already available and detailed personal preferences can be derived, supplementary consumption data may not have any (or a very limited) added value for the recommendation algorithm. These additional consumption data increase the required storage capacity and the computational load to generate the personal recommendations. Moreover, since personal preferences and the relevance of content items may vary over time, older consumption data might be outdated and lead to inaccurate recommendations. Therefore, we investigate which consumption data are (the most) relevant to feed the conventional collaborative filtering algorithms. For provider-generated content systems, we demonstrate that the accuracy of collaborative filtering algorithms increases by extending user profiles with additional older consumption data. In contrast, we witness the opposite effect for user-generated content systems: involving older consumption data has a negative influence on the recommender accuracy. These results are important for website owners who intend to employ a recommendation system at a minimum storage and computation cost.