A Rich Ranking Model Based on the Matthew Effect Optimization

Author(s):  
Jinzhong Li ◽  
Guanjun Liu
2018 ◽  
Vol 2018 ◽  
pp. 1-11 ◽  
Author(s):  
Jinzhong Li ◽  
Guanjun Liu

Matthew effect is a desirable phenomenon for a ranking model in search engines and recommendation systems. However, most of algorithms of learning to rank (LTR) do not pay attention to Matthew effect. LambdaMART is a well-known LTR algorithm that can be further optimized based on Matthew effect. Inspired by Matthew effect, we distinguish queries with different effectiveness and then assign a higher weight to a query with higher effectiveness. We improve the gradient in the LambdaMART algorithm to optimize the queries with high effectiveness, that is, to highlight the Matthew effect of the produced ranking models. In addition, we propose strategies of evaluating a ranking model and dynamically decreasing the learning rate to both strengthen the Matthew effect of ranking models and improve the effectiveness of ranking models. We use Gini coefficient, mean-variance, quantity statistics, and winning number to measure the performances of the ranking models. Experimental results on multiple benchmark datasets show that the ranking models produced by our improved LambdaMART algorithm can exhibit a stronger Matthew effect and achieve higher effectiveness compared to the original one and other state-of-the-art LTR algorithms.


1991 ◽  
Vol 19 (2) ◽  
pp. 253-259 ◽  
Author(s):  
John J. Horan ◽  
Chris D. Erickson

Graduates and employees of four institutions-the MOMM Cartel-dominate every science and practice organ of Division 17's governing body. Counseling psychologists from the other 60 academic programs (the OUTSIDERs) face numerous barriers to professional ascendancy. Six of 13 fellow nominees during the 1988-1989 year were MOMM members; none were rejected. Three OUTSIDERs were elected; four were rejected. Mean scholarly productivity for each group was 13, 21.3, and 19.7 Psyc LIT citations, respectively. The accepted OUTSIDERs were significantly more productive than the MOMMs; two of the four rejected OUTSIDERs ranked numbers one and two in scholarly productivity among all nominees. No relationship appeared between scholarship and fellow decisions; MOMM membership strongly predicted election to fellow status. Personal familiarity with the evaluators, rather than professional service, apparently accounts for these filings -a variant of the "Matthew Effect" discussed in the sociology-of-science literature Recommendations for reform are offered


2017 ◽  
Vol 27 (4) ◽  
pp. 362-364 ◽  
Author(s):  
Michael Sauder

Existing research on the Matthew Effect establishes that this dynamic can alter information flow and the distribution of rewards in ways that lead to cumulating advantages for high status actors. We know little, however, about how systems of evaluation, and especially variations in systems of evaluations, influence the expression and strength of these outcomes. Drawing on analyses of the effects of rankings on organizations, I consider how different evaluation contexts can change both audience perceptions about which organizations are award worthy and the definition of merit on which reward distributions are based.


Author(s):  
T.V. Rastimehina ◽  

The author examines the actions of states in crisis and emergency situations and conducts a comparative analysis of the emergency measures taken by the governments of democratic states and hybrid regime states. The author notes the Matthew effect: the actions of the institutions of political power of democratic and undemocratic countries in the conditions of the regime of increased readiness at a superficial examination seem similar, but have a different effect. In democracies, the restriction of some of the freedoms of citizens fits into the general trend toward humanization of politics. At the same time, autocracies presumably use the crisis to redistribute power resources toward the executive branch and to normalize the suppression of civic engagement.


2018 ◽  
Author(s):  
António Caleiro

A partial view of the Matthew effect in science asserts that the (already) most recognized scientists are those who (more easily) gain greater recognition for their scientific contributions. A full view of that effect naturally adds to the (comparative) advantages of the most recognized scientists, the (comparative) disadvantages of lesser-recognized scientists. The purpose of this report is to present one of the simplest explanations of the Matthew effect in science, which, as it is also very general, can explain the existence of that effect in other areas where inequality is manifested.


Physics Today ◽  
1991 ◽  
Vol 44 (10) ◽  
pp. 154-154 ◽  
Author(s):  
Douglas F. Brewer

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Rajesh V. Srivastava ◽  
Thomas Tang

Purpose In an ongoing War for Talent, what are the intangible and tangible return on investments (ROIs) for boundary-spanning employees? This study aims to develop a formative structural equation model (SEM) of the Matthew effect in talent. management. Design/methodology/approach This study develops a formative SEM theoretical model. Training and development (T&D) are the two antecedents of the latent construct – talent management strategy (TMS). This study frames the latent construct (TMS) in the proximal context of reducing burnout (cynicism and inefficacy), the distal context of subjective and intangible outcomes (job and life satisfaction) and the omnibus context of objective, tangible and financial rewards (the sales commission). The study collected data from multiple sources – objective sales commission from personnel records and subjective survey data from 512 sales employees. Findings The empirical discoveries support the theory. Both T&D contribute significantly to the TMS, which reduces burnout in the immediate context. TMS enhances job satisfaction more than life satisfaction in the distal context. TMS significantly and indirectly improves boundary spanners’ sales commission in the omnibus context via life satisfaction, but not job satisfaction. The model prevails for the whole sample, men, but not women. Practical implications Our discoveries offer practical implications for the Matthew effect in talent management: policymakers must cultivate T&D, develop TMS, facilitate the spillover effect from job satisfaction to life satisfaction, concentrate on the meaning in their lives and take their mind off money. TMS ultimately helps ignite these boundary spanners’ sales commission and their organization’s bottom line and financial health. The rich get richer. Originality/value It is life satisfaction (not job satisfaction) that excites boundary-spanning employees’ high level of sales commission. Our model prevails for the whole sample and men, but not for women. Job satisfaction spills over to life satisfaction for the entire sample, for men, but not for women. The results reveal gender differences.


Sign in / Sign up

Export Citation Format

Share Document