Search results

Many recommender systems are generally unable to provide accurate recommendations to users with limited interaction history, which is known as the cold-start problem. This issue can be resolved by trivial approaches that select random items or the most popular one to recommend to the new users. However, these methods perform poorly in many cases. This paper aims to explore the problem that how to make accurate recommendations for the new users in cold-start scenarios.

In this paper, the authors propose embedded-bandit method, inspired by Word2Vec technique and contextual bandit algorithm. The authors describe user contextual information with item embedding features constructed by Word2Vec. In addition, based on the intelligence measurement model in Crowd Science, the authors propose a new evaluation method to measure the utility of recommendations.

The authors introduce Word2Vec technique for constructing user contextual features, which improved the accuracy of recommendations compared to traditional multi-armed bandit problem. Apart from this, using this study’s intelligence measurement model, the utility also outperforms.

Improving the accuracy of recommendations during the cold-start phase can greatly raise user stickiness and increase user favorability, which in turn contributes to the commercialization of the app.

The algorithm proposed in this paper reflects that user contextual features can be represented by clicked items embedding vector.

The purpose of this paper is to research the effect of uncertain rewards on the recommendation intention in referral reward programs (RRPs) and investigate the interaction of tie strength and reward type on the recommendation intention.

The research adopts a quantitative exploratory approach through the use of experiments. Study 1 adopted a 2×2 between-participants design ((reward type: certain reward vs uncertain reward)×(tie strength: strong tie vs weak tie)). Respectively, by manipulating uncertain probabilities and expected value, Studies 2 and 3 further explore the effect of uncertain rewards and tie strength on customers’ referral intention.

This paper finds the following: compared to certain rewards, customers’ referral intention under uncertain rewards is higher and positive experience has a mediating effect between reward type and recommendation intention; when only the recommender is rewarded, the tie strength between the recommender and the receiver moderates the effect of reward type on the recommendation intention; for strong ties, customers’ recommendation intention is higher in uncertain reward condition, but for weak ties, customers’ willingness to recommend is almost the same in both reward types; when both the recommender and the receiver are rewarded, although certain rewards have a higher expected value than uncertain and random rewards, for strong ties, the participants have a higher referral intention under random rewards than that under uncertain rewards, which have a higher referral willingness than that under certain rewards. Additionally, for weak ties, the reverse is true.

The research has both theoretical implications for research on uncertain rewards and tie strength and practical implications for marketing managers designing and implementing RRPs.

Social media provide an impressive amount of data about users and their interactions, thereby offering computer and social scientists, economists, and statisticians – among others – new opportunities for research. Arguably, one of the most interesting lines of work is that of predicting future events and developments from social media data. However, current work is fragmented and lacks of widely accepted evaluation approaches. Moreover, since the first techniques emerged rather recently, little is known about their overall potential, limitations and general applicability to different domains. Therefore, better understanding the predictive power and limitations of social media is of utmost importance.

Different types of forecasting models and their adaptation to the special circumstances of social media are analyzed and the most representative research conducted up to date is surveyed. Presentations of current research on techniques, methods, and empirical studies aimed at the prediction of future or current events from social media data are provided.

A taxonomy of prediction models is introduced, along with their relative advantages and the particular scenarios where they have been applied to. The main areas of prediction that have attracted research so far are described, and the main contributions made by the papers in this special issue are summarized. Finally, it is argued that statistical models seem to be the most fruitful approach to apply to make predictions from social media data.

This special issue raises important questions to be addressed in the field of social media-based prediction and forecasting, fills some gaps in current research, and outlines future lines of work.

The MoS₂/graphite composite coatings modified by La₂O₃ through spraying technique were successfully prepared on the inner rings of spherical plain bearings. As a comparison, unmodified coatings were also prepared. This paper aims to study the La-modified MoS₂/graphite composite coating experimentally and improve the tribological performance of self-lubricating spherical plain bearings.

The performance of La₂O₃ toward the friction coefficient, temperature rise and wear rate of the coatings was studied by a self-made tribo-tester under different swing cycles. And the texture, surface morphology and element composition of the coatings were characterized by scanning electron microscope, energy dispersive spectroscopy and X-ray diffractometry.

The additives La₂O₃ refined the coatings’ microstructure and improved the tribological properties of the coatings. The oxidation of Mo + 4 to Mo + 6 was effectively inhibited. And the amount of abrasive grains, peeling pits and local cracks on the coatings surface decreased and homogeneous lubricating films formed, which were attributed to the existence of La₂O₃. The wear mechanisms of unmodified coatings were severe abrasive wear, adhesive wear and delamination wear. However, it exhibited superior wear resistance of the La-modified coatings to unmodified coatings, presenting slight abrasive wear and adhesive wear. The service life of bearings was prolonged under the protection of the modified coatings.

The paper proposed a new modified MoS₂/Graphite composite coating for the self-lubricating spherical plain bearings. The investigation on the friction, wear and temperature increase behaviors and the wear mechanisms of the coatings are beneficial to prolonging the service life of the self-lubricating spherical plain bearings.

Lower-limb exoskeleton systems enable people with spinal cord injury to regain some degree of locomotion ability, as the expected motion curve needs to adapt with changing scenarios, i.e. stair heights, distance to the stairs. The authors’ approach enables exoskeleton systems to adapt to different scenarios in stair ascent task safely.

In this paper, the authors learn the locomotion from predefined trajectories and walk upstairs by re-planning the trajectories according to external forces posed on exoskeleton systems. Moreover, instead of using complex sensors as inputs for re-planning in real-time, the approach can obtain forces acting on exoskeleton through dynamic model of human-exoskeleton system learned by an online machine learning approach without accurate parameters.

The proposed approach is validated in both simulation environment and a real walking assistance exoskeleton system. Experimental results prove that the proposed approach achieves better performance than the traditional predefined gait approach.

First, the approach obtain the external forces by a learned dynamic model of human-exoskeleton system, which reduces the cost of exoskeletons and avoids the heavy task of translating sensor input into actuator output. Second, the approach enables exoskeleton accomplish stair ascent task safely in different scenarios.

This paper aims at the problem of attaching the data of doctors, patients and the real-time sensor data of the exoskeleton to the cloud in intelligent rehabilitation applications. This study designed the exoskeleton cloud-brain platform and validated its safety assessment.

According to the dimension of data and the transmission speed, this paper implements a three-layer cloud-brain platform of exoskeleton based on Alibaba Cloud's Lambda-like architecture. At the same time, given the human–machine safety status detection problem of the exoskeleton, this paper built a personalized machine-learning safety detection module for users with the multi-dimensional sensor data cloned by the cloud-brain platform. This module includes an abnormality detection model, prediction model and state classification model of the human–machine state.

These functions of the exoskeleton cloud-brain and the algorithms based on it were validated by the experiments, they meet the needs of use.

This thesis innovatively proposes a cloud-brain platform for exoskeletons, beginning the digitalization and intelligence of the exoskeletal rehabilitation process and laying the foundation for future intelligent assistance systems.

This paper aims to explore the influence of CO₂ partial pressure, flow rate and water cut on N80 steel corrosion behaviors in the displacement process of oil in glutenite reservoir by CO₂ injection.

A self-made 3 L high-temperature and high-pressure autoclave was used to conduct corrosion simulation experiments of N80 steel in different CO₂ partial pressures, flow rates and water cut (the independently developed oil and water mixing approach can ensure the uniform mixing of oil and water in experiments). Techniques like weight loss and surface analysis were used to analyze the corrosion behaviors of N80 steel under different conditions.

Results showed that the average corrosion rate of N80 steel accelerated at varying degrees with the increase of CO₂ partial pressure, flow rate and water cut. Excluding that the samples showed uniform corrosion under the two conditions of 0.5MPa CO₂ partial pressure and static corrosion, they displayed mesa attack corrosion under other conditions. Besides, with the increase of CO₂ partial pressure, the pH value of solution dropped and the matrix corrosion speed rose, hence leading to the increased Fe²⁺ and CO₃²⁻ concentration. Meanwhile, a lowered pH value improved the FeCO₃ critical supersaturation, thereby leading to an increased nucleation rate/growth rate and ultimately causing the decrease of the dimension of FeCO₃ crystallites formed on the surface of the samples.

The results can be helpful in targeted anti-corrosion measures for CO₂/oil/water corrosive environment.

The objective of inventory management models is to determine efficient policies for managing the trade-off between customer satisfaction and the cost of goods. This chapter presents a methodology that uses the Monte Carlo Method (MCM) to estimate the behavior of a raw material supply model, considering uncertain variables such as demand, prices, and exchange rates. In order to show how to use this methodology, we analyze the case of a Colombian company in the aluminum industry. This company imports aluminum sheets from China. In this case, we analyze the financial impact of the raw material supply contract proposed by the Chinese supplier. The model considers different supply scenarios for the raw material. We calculate robust indicators such as Value at Risk (VaR), the Conditional Value at Risk (CVaR) and the probability of success for each scenario analyzed. Finally, we conduct a sensitivity analysis with respect to the sales price to validate the proposed models and solution approaches. The results show that considering risk metrics to evaluate the impact of endogenous factors over the supply process is a useful approach to improve decision-making related to this process and also can help to ensure the profitability of the company.

Studies in teacher research have seldom discussed the effects of writing publications and teachers’ views on it. This paper aims to examine how teachers in China regard the purpose of writing up research papers and its impact on their professional practice. It also investigates facilitating factors in preparing and writing publications.

This is a qualitative study which mainly analyses interview data gathered from 14 teachers, who have experience in writing publications, from three schools in Shanghai, China. Documentary analysis includes selected published journal papers and book chapters of the interviewees.

Findings show that teachers perceive producing publications as serving both an instrumental purpose of career advancement and a developmental purpose of enhancing their individual professional competency which in turn improves the quality of the teaching profession through knowledge transfer and knowledge transportation.

Contributing factors to help teachers to write include the provision of systematic training courses for teachers and the development of network and research partners.

Findings of this study help school leaders and teacher educators better their understanding of improving the quality of practice of teachers through equipping them with a “stance of inquiry”. Findings suggest ways to provide relevant support to enhance the writing capacity of teachers. More importantly, a revision of the existing teacher promotion system may make the practice of writing publications of more direct value to teachers’ daily practice.

This paper aims to examine the effect of consumer inertia – a tendency to adhere to prior purchase decisions despite the existence of preferable alternatives – on the relationships between switching costs and customer retention, and explain the controversy within them in the context of budget hotels.

An empirical test was conducted via structural equation modelling based on 337 self-administered questionnaires from customers at six of Chinese popular budget hotels.

Consumer inertia was a significant mediator between switching costs and customer retention. Specifically, consumer inertia completely mediated the effects of financial switching costs on customer retention, but partially did between procedural switching costs and customer retention. By introducing consumer inertia, the explanatory power of customer retention improved significantly by 19%.

This paper clarifies the effects of multiple dimensions of switching costs on customer retention via the mediating role of consumer inertia and partly addresses the inconsistency in the prior studies from an inertia perspective.