Identification of factors predicting clickthrough in Web searching using neural network analysis

Abstract

In this research, we aim to identify factors that significantly affect the clickthrough of Web searchers. Our underlying goal is determine more efficient methods to optimize the clickthrough rate. We devise a clickthrough metric for measuring customer satisfaction of search engine results using the number of links visited, number of queries a user submits, and rank of clicked links. We use a neural network to detect the significant influence of searching characteristics on future user clickthrough. Our results show that high occurrences of query reformulation, lengthy searching duration, longer query length, and the higher ranking of prior clicked links correlate positively with future clickthrough. We provide recommendations for leveraging these findings for improving the performance of search engine retrieval and result ranking, along with implications for search engine marketing.

To what degree can log data profile a web searcher?

“…The goal is to determine more efficient methods to optimize click through rate. In one study, we used a neural network to detect the significant influence of searching characteristics on future user click through (Zhang Y., Jansen, Spink, 2009a). Neural networks are powerful data modeling tools and are able to capture and represent complex relationships between input and output, so they have broad application.…”

Short-Term Effects of Fics and Cics Advertisements -Purchase Funnel Model Perspective

“…In practice, online advertising providers usually analyze the short-term instant effect of online advertisement through the click rate and purchase conversion rate of online advertisement on the day. Zhang et al used artificial neural network algorithm to model the user's click behavior of paid search ads, and found that the length of keywords in paid search ads and the content of advertisement itself have a great influence on user click ads [3]. Moe et al use the clickstream data of user ads to predict the user purchase conversion rate according to the user's historical visits and purchase records to evaluate the short-term effect of online ads [4].…”

Distributed lags using elastic-net regularization for market response models: focus on predictive and explanatory capacity

“…Some examples of works addressing sales-marketing responses are based on Bayesian regressions (Brown, 1986;Bass et al, 2007), Bayesian hierarchical methods, multivariate linear regressions (Havlena & Graham, 2004), the Kalman Filtering as is the case of Naik et al (1998), Support Vector Machines (SVM) (Viaene et al, 2001) Artificial Neural Networks (ANNs) (Zhang et al, 2009;Guido et al, 2011), nonlinear dynamic time series (Huffaker & Fearne, 2019), dynamic linear model combining multiple submodels for accounting for underlying patterns of advertising (Bruce et al, 2012), Distributed Lag Models (Bass & Clarke, 1972;Clarke, 1976;Weinberg & Weiss, 1982;Rufino, 2008;Mulchandani et al, 2019). Regrettably, it is the rule that these studies solely address in-sample evaluation and, in less often, discuss theoretical assumptions, so they fail to provide sufficient evidence of the validity of their suggestions to practitioners for applying such methods in real-life practice.…”