Towards a Tunable Framework for Recommendation Systems Based on Pairwise Preference Mining Algorithms

In Recommender Systems, recommendation tasks are usually implemented by preference mining. Most existing methods in context-aware recommendation focus on learning user interests from all the records to implement preference mining. However, some records contribute to recommendations, whereas some others may reduce the performances. To address these limitations, we propose a division learning strategy to divide the original records into several groups based on regression tree techniques. Then, a two-layer preference mining process is carried out to produce group and local preferences. Finally, the two preferences are integrated by the preference integration (Prin) approach to give recommendations. The experimental results demonstrated that our model outperformed other state-of-the-art methods, which illustrated the importance of targeted modeling.

This article addresses search engine personalization. We present a new approach to mining a user's preferences on the search results from clickthrough data and using the discovered preferences to adapt the search engine's ranking function for improving search quality. We develop a new preference mining technique called SpyNB , which is based on the practical assumption that the search results clicked on by the user reflect the user's preferences but does not draw any conclusions about the results that the user did not click on. As such, SpyNB is still valid even if the user does not follow any order in reading the search results or does not click on all relevant results. Our extensive offline experiments demonstrate that SpyNB discovers many more accurate preferences than existing algorithms do. The interactive online experiments further confirm that SpyNB and our personalization approach are effective in practice. We also show that the efficiency of SpyNB is comparable to existing simple preference mining algorithms.

Measuring the success possibility of implementing advanced manufacturing technology by utilizing the consistent fuzzy preference relations

Objective To explore current situation of depressive symptoms and analyze its influencing factors in stage 1-4 chronic kidney disease (CKD) patients. Methods From August 2014 to September 2014, a cross-sectional study was performed, and 214 outpatients chronic kidney disease (CKD) were recruited by convenience sampling method to evaluate depressive scores using the Zung self-rating depression scale in a grade A class three hospital of Sichuan province. Using single factor analysis method, a preliminary study was performed to explore the influencing factors of depression symptoms including gender, residence, education and the family income per capita. Results A total of 214 CKD depression scores was (47.98±10.38), which was significantly more than that of norm (t=20.472, P<0.01). Single factor analysis showed gender, residence, education and the family income per capita all influenced the depressive incidences (χ2=14.929, 5.413, 7.290, 9.917; P<0.05). Multiple stepwise regression analysis showed gender, residence, education and the family income per capita all were depression influencing factors (P<0.05). Conclusions Medical and health institutions should give great importance to depressive symptoms in CKD patients, especially for stage 1-4 CKD patients. Focus on assessment and early intervention can reduce the incidences of the depression and improve the prognosis. Key words: Chronic kidney disease; Depression; Investigation; Influencing factors

Purpose: The literature on interlocked directorate or interlocks has proposed the presence of cross-directors as an influential factor in the prestige of a firm (Mizruchi, 1996). This theory is based on the fact that, by having multiple companies with interlocks within them, major stakeholders of the company ought to have a higher perception of prestige. The aim of this preliminary study is to present an exploratory study and suggest future research lines on this issue. Design/methodology/approach: In a first preliminary study, we evaluate the interlocks network comprising companies belonging to the ranking established by the MERCO report in Spain for the year 2010. This is a well-known ranking of corporate reputation in Spain used in recent articles (Luna &amp; Fernandez, 2010). For further analysis, other variables are considered as being quoted in the Spanish market benchmark index, IBEX 35, and the location of the headquarters. Findings: Significant positive relationship was obtained between the number of directorships shared with other public companies and the perceived prestige that stakeholders perceive from the firm. Similarly, we obtain significant positive correlation between the prestige and being listed on the IBEX 35. Finally, we identified banking, construction and energy sectors as the ones that use interlocking directorates more intensively. Research limitations: The results of this preliminary study show the viability of future studies about prestige and interlocks (Mizruchi, 1996). For the validation of the theory described in this study, it should be replicated in different regions and, with larger samples, to observe the effect that interlocks have on the perception of prestige of the company. Practical implications / Originality/value: This is among the first studies showing the existence of a relationship between interlocks and business prestige. It highlights the importance of the composition of the boards for the perceived prestige of enterprises.

The performance of weapon systems is the important factor in deciding the outcome of war and expressed in ROC(Required Operational Capability). ROC is determined at the phase of requirement decision and has a decisive effect on the total life cycle of weapon systems including acquisition and its disposal phase. Improper ROC is very likely to cause the increase of acquisition & operating cost, total life cycle cost, and the delay of deployment, influencing factors on determination of ROC must be thoroughly studied. However, domestic academic research on influencing factors are rarely conducted, and most of the research is part of the research on requirement planning and acquisition, the research results provide only general suggestion. Thereby, this research intended to provide systematic and empirical analysis on what factors influence on determination of ROC and what their relative importance weights. To achieve above mentioned research purpose, first we reviewed relevant institutionalized systems in home and ROC modification status along with their reasons in home. Then, we verified and finalized influencing factors on ROC derived from preliminary research through Delphi Technique and Factor Analysis, and analyzed relative priorities through pair comparison by AHP method. From 2006 to 2017, on average 15 cases of modifying ROC occurred annually and needs of such modification appear to grow. Modification of operational concept of weapon systems comprises 31% of the reasons to modify of ROC, followed by correction of improper ROC is 17%, correction of excessive ROC is 14%, collateral changes by change of interoperable weapon systems' ROC is 14%, respectively. As for relative importance weight of finalized influencing factors, 'Weapon systems operational concept and mounting system', 'Weapon systems major performance elements' are found to be high. In lower level of 18 factors, the order of importance weight is as follows 'Weapon systems operational concept', 'Survivability', 'Enemy weapon systems performance' , 'Mounting system performance' and 'Existing similar weapon systems performance' and 'Interoperability'. The significance of this research would be extensive investigation on entire relevant minutes of meeting of ROC modification, chronological analysis of reason of such modification along with its status, and systematic analysis and relative importance weights of influencing factors on ROC.

Sociodemographic characteristics, including regional variations, have been associated with different food consumption patterns. Behavioral factors and lifestyle variables may also contribute to different food dietary trends. In this way, the present study intended to investigate the consumption habits of the most relevant types of dairy products around the world and relate them to sociodemographic factors, for example, age, sex, education and country as well as with some anthropometric and behavioral aspects, for example, body mass index, satisfaction with body weight and exercise or sedentary lifestyles. One other objective of the study was to categorize the lifestyles of the participants, according to measured variables linked with hours of inactivity or exercise, in order to use these as possible differentiating variables for the consumption of dairy products. The study involved a questionnaire survey undertaken on a non-probabilistic convenience sample of participants from Portugal (PT) and Brazil (BR), and participation was voluntary and anonymous. The data analysis involved different statistical techniques: basic statistics, chi-square tests, factor analysis, cluster analysis and tree classification analysis. The results showed that semi skimmed milk is never consumed by about half of the participants (47.4% for PT and 46.7 for BR), and those numbers increase for skimmed (64.8% for PT and 50.9% for BR), chocolate flavored milk (82.6% for PT and 65.6% for BR) and enriched milks (94.8% for PT and 85.3% for BR). Cheeses are also consumed in the two countries by small numbers of people. The number of participants consuming imported cheeses in both countries was particularly low (only 4.0% consume these more than once a week in both countries), suggesting national products may be preferred. It was further observed that those who consume cheese do it seldom (once a week) or sometimes (2–3 times per week). Butter is also consumed by only about half of the adult population (43.8% for PT and 49.5% for BR), but the percentage of those who never consume butter increases for skimmed butter (66.0% for PT and 82.6% for BR) and unsalted butter (70.2% for PT and 69.1% for BR). The consumption of yogurts also follows similar low consumption patterns. The most frequently consumed yogurt types in Portugal are liquid (30.5% consume regularly) and natural yogurts (34.8% consume regularly), while in Brazil the most frequent are creamy fruit pulp yogurt (14.4% consume regularly), liquid (13.7% consume regularly) and Greek type yogurt (10.2% consume regularly). A factor analysis and a cluster analysis established groups according to lifestyles, as follows: 1—Screeners, 2—Exercisers, 3—Travelers and 4—Others. These lifestyles were found to be influential in the consumption of dairy products for all classes of dairy tested: milk, cheese, yogurt and butter. For example, the screeners were found to consume more milk, more butter, more cheese and more yogurt. Additionally, other influential factors were age, sex, education, BMI and satisfaction with body weight. Nevertheless, country was not a meaningfully discriminant variable in relation to the other variables included in the classification analysis. The results concluded that, despite some small differences in the patterns of consumption of dairy products in both countries, the levels of consumption of dairy products are extremely low, for all classes studied (milk, cheese, yogurt or butter). Additionally, it was concluded that some factors are influential on the level of consumption of dairy products, and therefore decision makers can plan their interventions according to the characteristics of the targeted segments of the population, according to lifestyle, age, sex, education, BMI and satisfaction with body weight.

PurposeThis study aims to conduct an overview of previous studies on job satisfaction, particularly its determinants and outcomes, and the research objectives, main themes and years of publication of previous studies. This study also seeks to analyze research trends on job satisfaction in the field of hospitality and tourism.Design/methodology/approachThe top hospitality and tourism journals were reviewed, and relevant papers were searched using the keyword “job satisfaction.” Content analysis was performed to identify the research objectives, main themes, influencing factors, outcomes and journals.FindingsA total of 143 refereed journal papers were collected, of which 128 papers explored the influencing factors of job satisfaction, and 53 papers aimed to investigate outcomes. The predictors of job satisfaction were further classified into four groups, namely, organizational, individual, social and family and psychological factors.Research limitations/implicationsThis study conducted a literature review on job satisfaction by using content analysis. A relatively comprehensive review of job satisfaction is provided. However, this preliminary study still has considerable room for improvement given the extensive studies on job satisfaction. Future studies may perform meta-analysis and attempt to find new values of job satisfaction.Practical implicationsFindings may shed light on practical management. From the individual perspective, education, interest and skills were found to be related to job satisfaction. Thus, managers should provide their employees with opportunities to train and update their skills. From the organizational perspective, organizational support and culture contributed positively to job satisfaction. This perspective highlighted the importance of effective management activities and policies. From the social and family perspective, family–work supportive policies must be implemented to enhance job satisfaction. From the psychological perspective, psychological issues were found to be closely related to job satisfaction. Thus, the employees’ stress should be reduced to ensure that they perform their jobs well.Social implicationsThis study analyzed the determinants and outcomes of job satisfaction and highlighted the importance of enhancing job satisfaction from different perspectives. The interest of employees should be enhanced, their family–work conflict should be reduced and their psychological issues should be addressed to stimulate their enthusiasm. As job satisfaction contributes positively to organizational commitment and intention to stay, managers should conduct a series of organizational supportive activities to enhance job satisfaction, which will retain qualified employees.Originality/valueThis study conducted extensive research on job satisfaction and drew a systematic picture of job satisfaction on the basis of its determinants and outcomes, research objectives, main themes and journals. All findings were comprehensive and combined to contribute to the literature and serve as a foundation for further study.

Many young and middle-aged maintenance hemodialysis patients suffer a poor prognosis, experience a series of problems during long-term treatment and are thus prone to stigma. This study was designed to analyze stigma in young and middle-aged maintenance hemodialysis patients and explore its influencing factors. This study was conducted as a cross-sectional descriptive study with a convenience sampling method and included 97 patients from Shanghai Jiao Tong University Affiliated Sixth People’s Hospital between November 2020 and February 2021. The Social Impact Scale, a demographic questionnaire, and hemodialysis indicators were used in the investigation. Patient biochemical indexes from hemodialysis were compared. Young and middle-aged maintenance hemodialysis patients had a medium level of stigma. Patients who had low income, were younger, were male, had long-term hemodialysis and were unemployed had significantly higher stigma scores than other patients (P < 0.05). Age, gender, occupational status, annual household income and the duration of hemodialysis were found to be the main factors related to stigma in young and middle-aged maintenance hemodialysis patients by multiple regression analysis. Perceptions of hemodialysis-related stigma were common in our sample. Patients who had low income, were younger, were male, had long-term hemodialysis and were employed had a higher level of stigma, which deserves attention from clinical medical workers. Replication studies are needed to confirm these findings.

The concepts in domain ontology have similarities, and a good similarity computation method can improve the efficiency of information retrieval system and recommendation system. The current similarity computation methods of concepts in domain ontology do not take full account of the structural information carried by the concept nodes, and their performance are still room for improvement. On the basis of studying the existing similarity computation methods, we propose an improved similarity computation module of concepts in domain ontology which based on basic semantics and influencing factors (BISCM). BISCM takes the similarities of semantic distance, information content (IC) and attribute between concepts as the basic semantics to conduct the similarity computation, then considers 3 kinds of influencing factors to adjust the similarity and get the final similarity. The influencing factors include the level sequence, hierarchy depth and the coincidence degree of nodes. The experimental results show that BISCM perfects the factors needed to be considered in the concepts similarity computation; the Pearson correlation between the similarities computed by BISCM between expert experiences ups to 95.1% and BISCM has well adaptability to computing the similarities of concepts in domain ontology.

This empirical study examines how philosophical work on the sublime relates to contemporary psychological work on awe. We operationalized several aspects of the sublime drawing from prominent philosophical theories and analyzed them in relation to three different measures of awe: the modified Differential Emotions Scale (mDES), the awe sub-scale of the Dispositional Positive Emotion Scale (DPES), and the Awe Experience Scale (AWE-S). We carried out an Exploratory Factor Analysis on our items on the sublime. We found high correlations between these items and the measures of awe, especially with the self-loss and connectedness dimensions of the AWE-S. By operationalizing aspects of the sublime drawn from influential philosophical theories and comparing them with psychological measures of awe, we find a large degree of overlap between awe and the sublime, suggesting that these two literatures could inform one another.

&amp;lt;p&amp;gt;The development of ACMANT homogenization software started during the European COST HOME project, around 2010. Due to its excellent results in method comparison tests, the development of ACMANT has been being continuous since then. While its first version was applicable only to the homogenization of monthly temperature series, the later versions are applicable to a wide range of climatic variables and either for monthly or daily time series.&amp;lt;/p&amp;gt;&amp;lt;p&amp;gt;The operation of ACMANT is fast and automatic, and it is easy to use that even for large size datasets. The method can homogenize together time series of varied lengths, well tolerate data gaps, includes outlier filtering and infilling of data gaps (optional). ACMANT includes modern and effective statistical tools for the detection and removal of inhomogenities, such as step function fitting, bivariate detection for breaks of annual means and seasonal amplitudes (where applicable), ANOVA correction method and ensemble homogenization with varied pre-homogenization of neighbour series. For these properties, ACMANTv4 was the most accurate homogenization method in most method comparison tests of the Spanish MULTITEST project (https://doi.org/10.1175/JCLI-D-20-0611.1). In these tests, one important exception occurred, namely network mean trend errors were removed with significantly higher certainty by the Pairwise Homogenization Algorithm when approximately a half of the time series were affected with quasi synchronous breaks imitating concerted technical changes in the performance of climate observations. The most recent developments aiming the release of ACMANTv5 include the elimination of this drawback of ACMANT.&amp;lt;/p&amp;gt;&amp;lt;p&amp;gt;For ACMANTv5, a new break detection method has been developed, in which the combination of two time series comparison methods is applied. The new method contains both the use of composite reference series and pairwise comparisons, and in the detection with composite reference series the step function fitting is forced to include the breaks detected by pairwise comparisons. Another novelty of ACMANTv5 is that it gives options to use metadata in the homogenization procedure. The default operation mode of ACMANTv5 is still fully automatic, with or without the automatic use of a prepared metadata table. ACMANTv5 uses every date of the metadata list as a break indicator, and they are evaluated together with other indicators obtained by pairwise comparisons. Optionally, ACMANTv5 gives access to users to edit the list of detected breaks based on the pairwise detections of the first homogenization round. In the later steps of ACMANTv5 user intervention is not possible, but metadata may be considered by the automatic procedure also in the final estimation of break positions.&amp;amp;#160; &amp;amp;#160;&amp;amp;#160;&amp;amp;#160;&amp;lt;/p&amp;gt;

As e-commerce offers more and more choices for users, its structure becomes more and more complicated. Inevitably, it brings about the problem of information overload. The solution to this problem is an e-commerce personalized recommendation system using machine learning technology. People often seem confused when facing extensive information and cannot grasp the key points. This paper studies the personalized recommendation technology of e-commerce: deeply analyzes the related technologies and algorithms of the e-commerce recommendation system and proposes the latest architecture of the e-commerce recommendation system according to the current development status of the e-commerce recommendation system. The system recommends accuracy and real-time requirements and divides the system into two parts: offline mining and online recommendation and analyzes and implements the functions and technologies of each part. User-based recommender systems, collaborative filtering recommender systems, and content-based recommender systems are analyzed, respectively. The personalized recommendation cannot only quickly help customers find the required commodity information in a wide range of complex information but also can compare more commodity information to help customers to judge. However, the existing recommendation system has some problems such as the lack of recommendation personality, the reduced relevance of recommendation, and the poor timeliness of recommendation. Finally, a recommendation system that combines three recommendation algorithms is designed, and experiments are carried out. The newly designed recommendation system is compared with three different recommendation systems, and a summary and outlook are made. Based on the introduction of the relevant theories, characteristics, and mainstream technologies of personalized recommendation based on machine learning, this document presents a constructive example of a model based on the factors that influence personalized e-commerce information recommendations in the retail sector. Through questionnaire surveys, we analyze and design the influencing factors for consumers to purchase personalized products after the survey and build a project using state-of-the-art field learning techniques. Through the model to test the eight hypotheses proposed in this paper, the results show that customer income level, customer online shopping experience, commodity prices, product quality, recommendation relevance, credit evaluation, and service quality will have a significant positive impact on shopping willingness and ultimately affect the customer’s shopping behavior. e-commerce platform can use this influencing factor to establish personalized information recommendation service mode.

Introduction. The article examines the influence of various factors on the academic performance of university students. The purpose of the article is to identify the degree of influence of various factors on first-year university students’ academic performance taking into account their gender. The study is based on students’ self-assessment. Materials and Methods. The methodological basis of the research was the articles of Russian and foreign scholars devoted to the factors affecting the academic performance of university students (A. Alhadabi, H. W. Brink, Y. García, H. G. Hidajat, E. V. Krekhovets, F. Moradi, N. S. Nurieva, A. O. Ryazhkin, V. N. Shlyapnikov, D. R. Valeeva, S. D. Vaulin, S. V. Yaroshevskaya, I. Živčić-Bećirević and etc.). The empirical study was carried out at Tyumen State University. The sample consisted of 232 first-year students of various fields of study, including 101 males and 131 females. Results. The analysis of the scholarly literature on the research problem enabled the authors to identify the factors that influence the academic performance of university students and propose a classification of these factors. Based on students’ self-assessment, the degree of influence of each of the selected factors on their academic performance was determined. Significant differences in the assessment of the influence of a few factors on academic performance related to the students’ gender were revealed. According to the self-assessment of students, the competence of academic staff has the strongest influence on their academic performance. Within the framework of the proposed classification of factors, the most important for students was a group of factors related to their personal characteristics. Factors related to randomness and academic fraud were ranked among the least important. Intrinsic motivation as a factor influencing academic performance turned out to be more important than extrinsic positive and extrinsic negative motivation. Conclusions. The experimental data show that women, to a greater extent than men, associate their academic performance with external factors (teachers’ competence, learning environment, study facilities), or with their innate personality traits (intellectual abilities, memory characteristics and attention, emotional balance). For men, a more significant factor in academic performance is relying on their own strengths and expectations of success from their efforts. At the same time, their academic performance is more determined by the social environment at the university.

The paper explores the spatial spillover of carbon emissions and influencing factors based on spatial econometric model using panel series data over the period 2000-2012 in case of China. The results confirm that carbon emissions have spatial dependence, spatial spillover effects of provincial carbon emissions and influencing factors are obvious; Economic growth, industrial structure and export dependence coefficients are positive, technology progress on reducing carbon emissions growth effects is remarkable, energy prices effects failed to pass the test of significance. The spatial spillover effects of carbon emissions and related influencing factors among neighbor provinces must be considered when government departments formulate policies and development planning. The paper opens up new insights for China decrease China's carbon emissions from different dimension as a whole.