Search results

Learning to rank algorithms inherently faces many challenges. The most important challenges could be listed as high-dimensionality of the training data, the dynamic nature of Web information resources and lack of click-through data. High dimensionality of the training data affects effectiveness and efficiency of learning algorithms. Besides, most of learning to rank benchmark datasets do not include click-through data as a very rich source of information about the search behavior of users while dealing with the ranked lists of search results. To deal with these limitations, this paper aims to introduce a novel learning to rank algorithm by using a set of complex click-through features in a reinforcement learning (RL) model. These features are calculated from the existing click-through information in the data set or even from data sets without any explicit click-through information.

The proposed ranking algorithm (QRC-Rank) applies RL techniques on a set of calculated click-through features. QRC-Rank is as a two-steps process. In the first step, Transformation phase, a compact benchmark data set is created which contains a set of click-through features. These feature are calculated from the original click-through information available in the data set and constitute a compact representation of click-through information. To find most effective click-through feature, a number of scenarios are investigated. The second phase is Model-Generation, in which a RL model is built to rank the documents. This model is created by applying temporal difference learning methods such as Q-Learning and SARSA.

The proposed learning to rank method, QRC-rank, is evaluated on WCL2R and LETOR4.0 data sets. Experimental results demonstrate that QRC-Rank outperforms the state-of-the-art learning to rank methods such as SVMRank, RankBoost, ListNet and AdaRank based on the precision and normalized discount cumulative gain evaluation criteria. The use of the click-through features calculated from the training data set is a major contributor to the performance of the system.

In this paper, we have demonstrated the viability of the proposed features that provide a compact representation for the click through data in a learning to rank application. These compact click-through features are calculated from the original features of the learning to rank benchmark data set. In addition, a Markov Decision Process model is proposed for the learning to rank problem using RL, including the sets of states, actions, rewarding strategy and the transition function.

The landscape of search engine usage has evolved since the last known data were used to calculate click-through rate (CTR) values. The objective was to provide a replicable method for accessing data from the Google search engine using programmatic access and calculating CTR values from the retrieved data to show how the CTRs have changed since the last studies were published.

In this study, the authors present the estimated CTR values in organic search results based on actual clicks and impressions data, and establish a protocol for collecting this data using Google programmatic access. For this study, the authors collected data on 416,386 clicks, 31,648,226 impressions and 8,861,416 daily queries.

The results show that CTRs have decreased from previously reported values in both academic research and industry benchmarks. The estimates indicate that the top-ranked result in Google's organic search results features a CTR of 9.28%, followed by 5.82 and 3.11% for positions two and three, respectively. The authors also demonstrate that CTRs vary across various types of devices. On desktop devices, the CTR decreases steadily with each lower ranking position. On smartphones, the CTR starts high but decreases rapidly, with an unprecedented increase from position 13 onwards. Tablets have the lowest and most variable CTR values.

The theoretical implications include the generation of a current dataset on search engine results and user behavior, made available to the research community, creation of a unique methodology for generating new datasets and presenting the updated information on CTR trends. The managerial implications include the establishment of the need for businesses to focus on optimizing other forms of Google search results in addition to organic text results, and the possibility of application of this study's methodology to determine CTRs for their own websites.

This study provides a novel method to access real CTR data and estimates current CTRs for top organic Google search results, categorized by device.

This study aims to address the following research questions: Do the two types of service firms (individual or aggregator) have similar competitiveness on online search ads? How should the two types of service firms select optimal branded keywords to improve search performance? In addition, how do consumers’ search queries influence the service search performance of the two types of service firms?

In this study, the authors conduct an empirical analysis by building a two-stage choice modeling on the process of search engine ranking and consumer click-through decisions. The authors estimate the parameter coefficients and test the hypotheses using maximum likelihood estimation in the logistic regression model.

The empirical findings suggest that consumer response rates are highly dependent upon three aspects (service types, branded keyword strategy and consumer search query). First, the authors found that service aggregators receive greater consumer responses than individual service providers. Second, depending upon the various branded keyword strategies (e.g. generic vs branded, “within-type” vs “cross-type”) implemented by service aggregators or individual firms, the expected consumer responses could be quite different. Finally, customer’s search query, being either generic or branded, also has direct effect and interactive effect with service type on how consumers would response to the sponsored ads in the service search process.

The limitation of the research is twofold. First, conversion rate is not considered in the model estimation due to the nature of the data set. Second, the discussion about the keywords selection strategies is focusing on the hospitality industry. Future research shall further validate the generalizability into other industries.

First, given this competitive advantage, service aggregators should take an aggressive approach to adopting paid search strategy in acquiring new users and enhance its brand salience in the service ecosystem. Second, when considering other competitor’s brand names to include, if a firm is a service provider (e.g. hotel), a strategy that can help it receive higher consumer response would be to use “within-type” rather than “cross-type” branded keyword strategy. If a firm is a service aggregator, a better branded keyword strategy would be to use “across-type” instead of “within-type” approach. In addition, given that consumer’s brand awareness can influence the effectiveness of branded keyword strategy, online service search should target consumers in earlier stages of a decision journey.

The authors believe their theoretical framework can provide actionable solutions to service firms to ease customer’s search process, increase customer’s stickiness using search engines and add value to the customer relationships with all services entities within the digital ecosystem.

This study is the first to expand online search marketing into granule examinations (main and interactive effects of three key factors) in the service search domain. First, the authors differentiate service firms into two categories – online travel aggregators and individual hotels in the model. Second, the authors introduce two sets of new classifications of branded keywords for online service search research (i.e. own versus other brand and “cross-type” versus “within-type” branded keywords). Third, this study integrates service consumers’ search word specificity into the conceptual framework which is often missing in previous online search research.

Belkin and Robertson (1976a) reflected on the ethical implications of theoretical research in information science and warned that there was potential for abuse of knowledge gained by undertaking such research and applying it to information systems. In particular, they identified the domains of advertising and political propaganda that posed particular problems. The purpose of this literature review is to revisit these ideas in the light of recent events in global information systems that demonstrate that their fears were justified.

The authors revisit the theory in information science that Belkin and Robertson used to build their argument, together with the discussion on ethics that resulted from this work in the late 1970s and early 1980s. The authors then review recent literature in the field of information systems, specifically information retrieval, social media and recommendation systems that highlight the problems identified by Belkin and Robertson.

Information science theories have been used in conjunction with empirical evidence gathered from user interactions that have been detrimental to both individuals and society. It is argued in the paper that the information science and systems communities should find ways to return control to the user wherever possible, and the ways to achieve this are considered.

The ethical issues identified require a multidisciplinary approach with research in information science, computer science, information systems, business, sociology, psychology, journalism, government and politics, etc. required. This is too large a scope to deal with in a literature review, and we focus only on the design and implementation of information systems (Zimmer, 2008a) through an information science and information systems perspective.

The authors argue that information systems such as search technologies, social media applications and recommendation systems should be designed with the recipient of the information in mind (Paisley and Parker, 1965), not the sender of that information.

Information systems designed ethically and with users in mind will go some way to addressing the ill effects typified by the problems for individuals and society evident in global information systems.

The authors synthesize the evidence from the literature to provide potential technological solutions to the ethical issues identified, with a set of recommendations to information systems designers and implementers.

This chapter presents a summary of the literature on the influence of the Internet and other digital innovations on markets, consumers, and firms. The review leads to a list of topics in need of research in the general areas of big data, online and mobile advertising, consumer search, online privacy, online reviews, social networks, platforms for online transactions, and the impact of the Internet on retail markets, including multi-channel and omni-channel retailing. We discuss the big data approaches that have been applied to problems of targeting and positioning and suggest areas for further development of these approaches. We also discuss the emerging area of mobile advertising, which can further enhance targeting. On the consumer side, the evidence indicates that the Internet has greatly lowered the costs of search and access to retailers. Much of the consumer data are transmitted to sellers, and much of the online advertising is transmitted to consumers, through platforms, such as Google. We conclude that better models of competition among these platform firms are needed and that they need to be examined for anti-trust violations. While online retailing has grown rapidly, it still has a relatively small share of retail sales. Since sellers can combine the advantages of online and offline channels, it has been common for sellers to branch into multi-channel retailing. Given the increased availability of detailed consumer data, omni-channel selling, which emphasizes strategies for the various touchpoints that lead to a transaction, is an area for further development.

The purpose of this study is to determine what the history of research in marketing implies for the reaction of the field to recent developments in technology due to the internet and associated developments.

This paper examines the introduction of new research topics over 10-year intervals from 1960 to the present. These provide the basic body of knowledge that drives the field at the present time.

While researchers have always borrowed techniques, they have refined them to make them applicable to marketing problems. Moreover, the field has always responded to new developments in technology, such as more powerful computers, scanners and scanner data, and the internet with a flurry of research that applies the technologies.

Marketing will adapt to changes brought on by the internet, increased computer power and big data. While the field faces competition for other disciplines, its established body of knowledge about solving marketing problems gives it a unique advantage.

This paper traces the history of academic marketing from 1960 to the present to show how major changes in the field responded to changes in computer power and technology. It also derives implications for the future from this analysis.

El objetivo de este estudio es examinar qué implica la historia de la investigación académica en marketing en la reacción del campo de conocimiento a los recientes desarrollos tecnológicos como consecuencia de la irrupción de Internet.

Esta investigación analiza la introducción de nuevos temas de investigación en intervalos de diez años desde 1960 hasta la actualidad. Estos periodos proporcionan el cuerpo de conocimiento básico que conduce al ámbito del marketing hasta el presente.

Aunque los investigadores tradicionalmente han tomado prestadas ciertas técnicas, las han ido refinando para aplicarlas a los problemas de marketing. Además, el ámbito del marketing siempre ha respondido a los nuevos desarrollos tecnológicos, más poder de computación, datos de escáner o el desarrollo de Internet, con un amplio número de investigaciones aplicando tales tecnologías.

El marketing se adaptará a los cambios provocados por Internet, aumentando el poder de computación y el big data. Aunque el marketing se enfrenta a la competencia de otras disciplinas, su sólido cuerpo de conocimiento orientado a la resolución de problemas le otorga una ventaja diferencial única.

Describe la historia académica del marketing desde 1960 hasta la actualidad, para mostrar cómo los principales cambios en este campo respondieron a los cambios tecnológicos. Se derivan interesantes implicaciones para el futuro.

Historia, Revisión, Cambio, Tecnología, Conocimiento, Internet, Datos, Métodos

Revisión general

Geographical query formulation is one of the key difficulties for users in search engines. The purpose of this study is to improve geographical search by proposing a novel geographical query reformulation (GQR) technique using a geographical taxonomy and word senses.

This work introduces an approach for GQR, which combines a method of query components separation that uses GeoNames, a technique for reformulating these components using WordNet and a geographic taxonomy constructed using the latent semantic analysis method.

The proposed approach was compared to two methods from the literature, using the mean average precision (MAP) and the precision at 20 documents (P@20). The experimental results show that it outperforms the other techniques by 15.73% to 31.21% in terms of P@20 and by 17.81% to 35.52% in terms of MAP.

According to the experimental results, the best created taxonomy using the geographical adjacency taxonomy builder contains 7.67% of incorrect links. This paper believes that using a very big amount of data for taxonomy building can give better results. Thus, in future work, this paper intends to apply the approach in a big data context.

Despite this, the reformulation of geographical queries using the new proposed approach considerably improves the precision of queries and retrieves relevant documents that were not retrieved using the original queries. The strengths of the technique lie in the facts of reformulating both thematic and spatial entities and replacing the spatial entity of the query with terms that explain the intent of the query more precisely using a geographical taxonomy.

Expert and novice readers tag documents with different descriptions; this study is intended to discover which readers would generate the most reliable and most representative sets of tags.

One group of experts and one group of novices were recruited. These two groups were asked to provide tags for document bookmarks in a Mozilla Firefox browser. In the experimental analysis we defined two measures – similarity and relevance – to describe the differences between the two groups.

Tags chosen by experts yielded better similarity and relevance values in all analyses. Tags chosen by the expert group had higher commonality in pairwise similarity analysis; moreover, the relevance analysis showed that tags chosen by experts reflected better understanding of the content.

Tagging behavior has become highly popular on the web, and its study has commercial merit. Tags from experts represent the structure behind the knowledge involved; expert representation may be vastly more helpful than novice representation for promoting understanding of content in an era characterized by an explosion of information.

Revenue maximization through improving click‐throughs is of great importance for price comparison shopping services (PCSSs) whose revenues directly depend on the number of click‐throughs of items in their itemsets. The purpose of this paper is to present an approach aiming to maximize the revenue of a PCSS by proposing effective itemset construction methods that can maximize the click‐throughs.

The authors suggest three itemset construction methods, namely naïve method (NM), exhaustive method (EM), and local update method (LM). Specifically, NM searches for the best itemset for an item in terms of textual similarity between an item and an itemset, while EM produces the best itemset for each item for maximizing click‐throughs by considering all the possible memberships of the item. Finally, through combining NM and EM, the authors propose an LM that attempts to improve click‐throughs by locally updating the memberships of items according to their ranks in each itemset.

Through evaluation of the proposed methods based on a real‐world dataset, it has been found that improvement of click‐throughs is small when itemsets are constructed by using the textual similarity alone. However, significant improvement in the number of click‐throughs was achieved when considering items' membership updates dynamically.

Unlike the previous studies that mainly focus on the textual similarity, the authors attempt to maximize the revenue through constructing itemsets that can result in more click‐throughs. By using the proposed methods, it is expected that PCSSs will be able to automatically construct itemsets that can maximize their revenues without the need for manual task.

Internet referral services are a common form of online marketing operating activities. To incentivize infomediaries and improve referral performance, brand retailers typically apply the cost-per-click (CPC) or the cost-per-sale (CPS) payments. The purpose of this paper is to investigate the effect of referral services on the optimal contract with CPC or CPS payments.

This paper studies a mechanism design problem for internet referral services. To maximize the expected utility of the brand retailer, an uncertain contract model is established in which the brand retailer's assessment of the infomediary's referral service capability is characterized as an uncertain variable. Then equivalent models under CPC and CPS payments are presented to obtain the optimal solutions.

The results demonstrate that under CPC payments, as the referral service capability increases, the optimal sales volume is increasing, and the optimal transfer payment first shows a declining and then a rising trend. The brand retailer is less likely to raise the optimal transfer payment for the infomediary given a higher CPC revenue-sharing fee percentage, which is counterintuitive. Under CPS payments, the optimal sales volume and transfer payment are also increasing in the referral service capability. In addition, an increase in the click-through rate leads to the infomediary's incremental marginal utility.

The value of this research is its application of incentive contracts to the internet referral services considering CPC or CPS payments. The results of this research can serve as a guide for retailers and infomediaries in their decision-making around online retailing.