Product Ranking on Online Platforms

2022 ◽  
Author(s):  
Mahsa Derakhshan ◽  
Negin Golrezaei ◽  
Vahideh Manshadi ◽  
Vahab Mirrokni
Keyword(s):  
Market Share ◽  
Synthetic Data ◽  
Consumer Welfare ◽  
Sequential Search ◽  
Consideration Set ◽  
Performance Guarantees ◽  
Ranking Algorithms ◽  
Online Platforms ◽  
Optimal Polynomial ◽  
Preference Weights

On online platforms, consumers face an abundance of options that are displayed in the form of a position ranking. Only products placed in the first few positions are readily accessible to the consumer, and she needs to exert effort to access more options. For such platforms, we develop a two-stage sequential search model where, in the first stage, the consumer sequentially screens positions to observe the preference weight of the products placed in them and forms a consideration set. In the second stage, she observes the additional idiosyncratic utility that she can derive from each product and chooses the highest-utility product within her consideration set. For this model, we first characterize the optimal sequential search policy of a welfare-maximizing consumer. We then study how platforms with different objectives should rank products. We focus on two objectives: (i) maximizing the platform’s market share and (ii) maximizing the consumer’s welfare. Somewhat surprisingly, we show that ranking products in decreasing order of their preference weights does not necessarily maximize market share or consumer welfare. Such a ranking may shorten the consumer’s consideration set due to the externality effect of high-positioned products on low-positioned ones, leading to insufficient screening. We then show that both problems—maximizing market share and maximizing consumer welfare—are NP-complete. We develop novel near-optimal polynomial-time ranking algorithms for each objective. Further, we show that, even though ranking products in decreasing order of their preference weights is suboptimal, such a ranking enjoys strong performance guarantees for both objectives. We complement our theoretical developments with numerical studies using synthetic data, in which we show (1) that heuristic versions of our algorithms that do not rely on model primitives perform well and (2) that our model can be effectively estimated using a maximum likelihood estimator. This paper was accepted by Gabriel Weintraub, revenue management and market analytics.

scholarly journals From ranking algorithms to ‘ranking cultures’

2018 ◽  
Vol 24 (1) ◽  
pp. 50-68 ◽  
Author(s):  
Bernhard Rieder ◽  
Ariadna Matamoros-Fernández ◽  
Òscar Coromina
Keyword(s):  
Qualitative Analysis ◽  
Social Power ◽  
Causal Explanation ◽  
Search Results ◽  
Ranking Algorithms ◽  
Online Platforms ◽  
Change Metrics ◽  
Empirical Approaches ◽  
Over Time

Algorithms, as constitutive elements of online platforms, are increasingly shaping everyday sociability. Developing suitable empirical approaches to render them accountable and to study their social power has become a prominent scholarly concern. This article proposes an approach to examine what an algorithm does, not only to move closer to understanding how it works, but also to investigate broader forms of agency involved. To do this, we examine YouTube’s search results ranking over time in the context of seven sociocultural issues. Through a combination of rank visualizations, computational change metrics and qualitative analysis, we study search ranking as the distributed accomplishment of ‘ranking cultures’. First, we identify three forms of ordering over time – stable, ‘newsy’ and mixed rank morphologies. Second, we observe that rankings cannot be easily linked back to popularity metrics, which highlights the role of platform features such as channel subscriptions in processes of visibility distribution. Third, we find that the contents appearing in the top 20 results are heavily influenced by both issue and platform vernaculars. YouTube-native content, which often thrives on controversy and dissent, systematically beats out mainstream actors in terms of exposure. We close by arguing that ranking cultures are embedded in the meshes of mutually constitutive agencies that frustrate our attempts at causal explanation and are better served by strategies of ‘descriptive assemblage’.

1. Competition policy and economics

2018 ◽  
Author(s):  
Richard Whish ◽  
David Bailey
Keyword(s):  
Market Power ◽  
Market Share ◽  
Competition Policy ◽  
Competition Law ◽  
Consumer Welfare ◽  
Market Definition ◽  
Economic Context ◽  
Competition Laws ◽  
Competition Process

This chapter provides an overview of competition law and its economic context. Section 2 describes the practices that competition laws attempt to control in order to protect the competition process. Section 3 examines the theory of competition and gives an introductory account of why the effective enforcement of competition law is thought to be beneficial for consumer welfare. Section 4 considers the expected functions of a system of competition law. Section 5 then introduces two key economic concepts, market definition and market power, that are important to a better understanding of competition policy. The chapter concludes with a table of market share figures that are significant in the application of EU and UK competition law.

scholarly journals Learning quantized neural nets by coarse gradient method for nonlinear classification

2021 ◽  
Vol 8 (3) ◽  
Author(s):  
Ziang Long ◽  
Penghang Yin ◽  
Jack Xin
Keyword(s):  
Neural Networks ◽  
Loss Function ◽  
Ad Hoc ◽  
Gradient Methods ◽  
Synthetic Data ◽  
Neural Nets ◽  
Theoretical Understanding ◽  
Performance Guarantees ◽  
Almost Everywhere ◽  
Gradient Based

AbstractQuantized or low-bit neural networks are attractive due to their inference efficiency. However, training deep neural networks with quantized activations involves minimizing a discontinuous and piecewise constant loss function. Such a loss function has zero gradient almost everywhere (a.e.), which makes the conventional gradient-based algorithms inapplicable. To this end, we study a novel class of biased first-order oracle, termed coarse gradient, for overcoming the vanished gradient issue. A coarse gradient is generated by replacing the a.e. zero derivative of quantized (i.e., staircase) ReLU activation composited in the chain rule with some heuristic proxy derivative called straight-through estimator (STE). Although having been widely used in training quantized networks empirically, fundamental questions like when and why the ad hoc STE trick works, still lack theoretical understanding. In this paper, we propose a class of STEs with certain monotonicity and consider their applications to the training of a two-linear-layer network with quantized activation functions for nonlinear multi-category classification. We establish performance guarantees for the proposed STEs by showing that the corresponding coarse gradient methods converge to the global minimum, which leads to a perfect classification. Lastly, we present experimental results on synthetic data as well as MNIST dataset to verify our theoretical findings and demonstrate the effectiveness of our proposed STEs.

scholarly journals Error curves for evaluating the quality of feature rankings

2020 ◽  
Vol 6 ◽  
pp. e310
Author(s):  
Ivica Slavkov ◽  
Matej Petković ◽  
Pierre Geurts ◽  
Dragi Kocev ◽  
Sašo Džeroski
Keyword(s):  
Predictive Models ◽  
Synthetic Data ◽  
Benchmark Problems ◽  
Feature Ranking ◽  
Ranking Algorithm ◽  
Target Variable ◽  
Ranking Algorithms ◽  
Error Measures ◽  
Nested Sets

In this article, we propose a method for evaluating feature ranking algorithms. A feature ranking algorithm estimates the importance of descriptive features when predicting the target variable, and the proposed method evaluates the correctness of these importance values by computing the error measures of two chains of predictive models. The models in the first chain are built on nested sets of top-ranked features, while the models in the other chain are built on nested sets of bottom ranked features. We investigate which predictive models are appropriate for building these chains, showing empirically that the proposed method gives meaningful results and can detect differences in feature ranking quality. This is first demonstrated on synthetic data, and then on several real-world classification benchmark problems.

scholarly journals The first glance at the growing impact of private labels on the upstream local food suppliers in Finland

2016 ◽  
pp. 1-8
Author(s):  
Xing Liu ◽  
Jyrki Niemi
Keyword(s):  
Market Share ◽  
Bargaining Power ◽  
Consumer Welfare ◽  
European Countries ◽  
Private Label ◽  
Private Labels ◽  
Close Link ◽  
Eu Countries ◽  
Concentration Levels ◽  
The Eu

The last twenty years has witnessed substantial changes in retailing across most European countries. Particularly, the increased prevalence of category management, the significant investment in new technology and improved logistics have enabled the supermarkets in the European countries acquired an increasing share of grocery markets. As a strategic tool, private labels (PL) has become increasingly important for European retailers to gain market share, loyalty of customers and reinforce the bargaining power toward suppliers and countervailing power against manufacturing brands. (Bonfrer and Chintagunta, 2004; Hansen et al., 2006; Groznik et. al. 2010 EU commission, 2011). The combination of recession and a retail food price spike during the last 5 years provides even more opportunity for PL growth as increasingly price-sensitive customers shift to PL alternatives.1(Volpe, 2011) According to statistics from Private Label Manufacturer Association (PLMA), the market share of PLs accounts for 17 to 48% of the groceries market in the EU in 2012. Many consumers see private label not only a trade-down but more often as another branded options. (Nielsen, 2010) In Finland, the sales of private labels have been growing significantly during the last five years. However, the total share of the sales is still lower than in the EU countries on average. PL share is commonly positively correlated to concentration levels in food retail. (Lincoln and Thomason, 2009). Table 1 presents the concentration of national grocery markets in a number of EU countries versus the market shares of PLs based on the volumes obtained from PLMA. Figure 1 displays the total market share of PLs including food and non-food in Finland calculated in value. Clearly, Finnish grocery trade is the most concentrated amongst EU members of states. Even though the market share of PLs in Finland has not reached as high level as the other European countries such as Germany and UK, the market share of PLs in food sector based on sales value, has been steadily grown from 7.6% in 2003 to 12% in 20123(See Figure 1). Given the close link between concentration levels and PL share, the expectation that PL market share in Finland is projected to increase by between 3- 5% points yearly in the coming five years. Compared to the current level of 12 percent, this entails that PL market share is set to over 20% in value in the coming 5 years. The growing importance of PLs has spawned an academic literature empirically investigating the factors that facilitates its success (Cotterill et al, 2000; Chintagunta et al, 2002; Richards et. al., 2007; USDA, 2011) different countries (Cordeiro, 2007; Kilic and Hakan, 2009). One of common consequences of high concentration and growing PLs sales is a growing imbalance of bargaining power within food supply chains, i.e. the power of supermarkets over their suppliers. In Finland, the economic and social effects of such imbalanced bargaining power on producers and processors are increasingly recognized (FCA, 2012)4, however, the empirical research related to PLs has been very limited (Delvecchoio, 2001). Amongst the limited publication related PLs, many concentrated on retailers and consumer’ welfare being (Gabrielsen and Sorgard, 2007; Perrin 2006; Uusitalo and Rökman 2004; The Economist Intelligence Unit: Industry report, 2010). However, very limited research (Suvanto et.al, 2006), stood into suppliers’ shoes.

scholarly journals AN IMPACT OF DIGITALIZATION ON COMPETITION POLICY: WORLD’S PRACTICE AND UKRAINE

2019 ◽  
pp. 33-45
Author(s):  
Vasylyna Kolosha
Keyword(s):  
Market Share ◽  
Consumer Choice ◽  
Competition Policy ◽  
Network Effects ◽  
Personal Data ◽  
Consumer Welfare ◽  
Relevant Market ◽  
Digital Platforms ◽  
Scale Network ◽  
Almost All

The article examines an impact of digitalization on modern competition policy. Author proves that a necessity to modify traditional instruments of competition policy is caused by such special features of economic rivalry at the digital markets as significant return on scale, network effects and growing importance of data. Author argues that competition authorities face challenges caused by digitalization at the almost all stages of determining of firm’s dominant position, especially in the case of determining of relevant market, its participants and estimation of their market share. Special issues of the activity of digital platforms as a key competition subjects at modern markets are analyzed. Author proves that in the most cases it is appropriate to consider each side of the platform as a separate market when the relevant market is determined. Determination of a single market of intermediation services is justified only if the single price is set for all platform clients and if there is the same degree of substitution of the services for each consumer group. The article shows that it is necessary to consider positive cross-group effect when SSNІP-test is used for economic analysis of digital platforms activity. The problems of evaluation of business platforms market share are exposed. Author argues that usage of revenues as a basis for market share calculation is not appropriate if platform sets zero-price. In this case market share should be calculated based on the number of user or intensity of usage. It is proven that the main criterion of competition policy efficiency – consumer welfare – should include not only the price but also such parameters as privacy, consumer choice, protection of personal data, switching costs. A necessity of modification of Ukrainian competition policy in response to challenges of digital era is proven.

PENGARUH KUALITAS PELAYANAN DAN PROMOSI TERHADAP PERLUASAN PANGSA PASAR PRODUK DANA TALANGAN HAJI DI BANK SYARIAH MANDIRI KCP UJUNGBERUNG DAN KCP RANCAEKEK BANDUNG

2018 ◽  
Vol 2 (1) ◽  
pp. 1-15
Author(s):  
Intan Nurrachmi
Keyword(s):  
Quality Of Service ◽  
Service Quality ◽  
Market Share ◽  
Significant Influence ◽  
Strong Correlation ◽  
T Test ◽  
Market Expansion ◽  
Difference Constraints ◽  
Branch Office

This study departs from the hajj bailout financing facility which is a booming product because of the customer's interest, but in this case there is a difference in the target achievement between Bank Syariah Mandiri (BSM) Ujungberung KCP which is less successful in improving the hajj bailout products while the Rancaekek KCP is very superior in one consolidation Ahmad Yani Branch Office Bandung. This is what is interesting for researchers to carry out this research, the difference constraints include service quality and promotion factors. This phenomenon raises problems that must be examined, namely how the influence of service quality and promotion of market share expansion products hajj bailouts at Bank Syariah Mandiri KCP Ujungberung and KCP Rancaekek Bandung. This study aims academically to contribute in the study of Islamic economics in worksheets, especially the quality of service and promotion of market share expansion and practically expected to be able to provide input to all employees of BSM KCP Ujungberung regarding the quality of service and promotion of market expansion of bailout products. Hajj that has been successfully carried out by BSM KCP Rancaekek.The conclusion of this study is that there is a significant influence of service quality on the expansion of market share by 53.3% with a strong correlation of 0.730 and through t test, where t counts at 8.245 (> t table), then H_0 is rejected and H_i is accepted. Furthermore, there is a significant influence of promotion on the expansion of market share by 30.3% with a moderate / sufficient correlation of 0.550 through t test, where t counts is 4.219 (> t table), then H_ (0) is rejected and H_i is accepted. Then there is a significant influence of service quality and promotion simultaneously to the expansion of market share by 60.6% and a strong correlation of 0.784 and through Test F, where F count is 67.023 (> F table), then 〖H〗 _ ( 0) rejected and H_i accepted.

Sign in / Sign up

Export Citation Format

Share Document