The real-time bidding (RTB), aka programmatic buying, has recently become the fastest growing area in online advertising. Instead of bulking buying and inventory-centric buying, RTB mimics stock exchanges and utilises computer algorithms to automatically buy and sell ads in real-time; It uses per impression context and targets the ads to specific people based on data about them, and hence dramatically increases the effectiveness of display advertising. In this paper, we provide an empirical analysis and measurement of a production ad exchange. Using the data sampled from both demand and supply side, we aim to provide first-hand insights into the emerging new impression selling infrastructure and its bidding behaviours, and help identifying research and design issues in such systems. From our study, we observed that periodic patterns occur in various statistics including impressions, clicks, bids, and conversion rates (both post-view and post-click), which suggest time-dependent models would be appropriate for capturing the repeated patterns in RTB. We also found that despite the claimed second price auction, the first price payment in fact is accounted for 55.4% of total cost due to the arrangement of the soft floor price. As such, we argue that the setting of soft floor price in the current RTB systems puts advertisers in a less favourable position. Furthermore, our analysis on the conversation rates shows that the current bidding strategy is far less optimal, indicating the significant needs for optimisation algorithms incorporating the facts such as the temporal behaviours, the frequency and recency of the ad displays, which have not been well considered in the past.

/pdf/real-time-bidding-for-online-advertising-measurement-and-1gs1ns7iw1.pdf

Real-time bidding for online advertising: measurement and analysis

Internet advertising is a fast growing business which has proved to be significantly important in digital economics. It is vitally important for both web search engines and online content providers and publishers because web advertising provides them with major sources of revenue. Its presence is increasingly important for the whole media industry due to the influence of the Web. For advertisers, it is a smarter alternative to traditional marketing media such as TVs and newspapers. As the web evolves and data collection continues, the design of methods for more targeted, interactive, and friendly advertising may have a major impact on the way our digital economy evolves, and to aid societal development. Towards this goal mathematically well-grounded Computational Advertising methods are becoming necessary and will continue to develop as a fundamental tool towards the Web. As a vibrant new discipline, Internet advertising requires effort from different research domains including Information Retrieval, Machine Learning, Data Mining and Analytic, Statistics, Economics, and even Psychology to predict and understand user behaviours. In this paper, we provide a comprehensive survey on Internet advertising, discussing and classifying the research issues, identifying the recent technologies, and suggesting its future directions. To have a comprehensive picture, we first start with a brief history, introduction, and classification of the industry and present a schematic view of the new advertising ecosystem. We then introduce four major participants, namely advertisers, online publishers, ad exchanges and web users; and through analysing and discussing the major research problems and existing solutions from their perspectives respectively, we discover and aggregate the fundamental problems that characterise the newly-formed research field and capture its potential future prospects.

/pdf/internet-advertising-an-interplay-among-advertisers-online-41j6a81c3s.pdf

Internet Advertising: An Interplay among Advertisers, Online Publishers, Ad Exchanges and Web Users

With the emergence of web-based social and information applications, entity similarity search in information networks, aiming to find entities with high similarity to a given query entity, has gained wide attention. However, due to the diverse semantic meanings in heterogeneous information networks, which contain multi-typed entities and relationships, similarity measurement can be ambiguous without context. In this paper, we investigate entity similarity search and the resulting ambiguity problems in heterogeneous information networks. We propose to use a meta-path-based ranking model ensemble to represent semantic meanings for similarity queries, exploit the possibility of using using user-guidance to understand users query. Experiments on real-world datasets show that our framework significantly outperforms competitor methods.

/pdf/user-guided-entity-similarity-search-using-meta-path-vdsm9yxtp3.pdf

User guided entity similarity search using meta-path selection in heterogeneous information networks

Discovering similar users with respect to their habits plays an important role in a wide range of applications, such as collaborative filtering for recommendation, user segmentation for market analysis, etc. Recently, the progressing ability to sense user contexts of smart mobile devices makes it possible to discover mobile users with similar habits by mining their habits from their mobile devices. However, though some researchers have proposed effective methods for mining user habits such as behavior pattern mining, how to leverage the mined results for discovering similar users remains less explored. To this end, we propose a novel approach for conquering the sparseness of behavior pattern space and thus make it possible to discover similar mobile users with respect to their habits by leveraging behavior pattern mining. To be specific, first, we normalize the raw context log of each user by transforming the location-based context data and user interaction records to more general representations. Second, we take advantage of a constraint-based Bayesian Matrix Factorization model for extracting the latent common habits among behavior patterns and then transforming behavior pattern vectors to the vectors of mined common habits which are in a much more dense space. The experiments conducted on real data sets show that our approach outperforms three baselines in terms of the effectiveness of discovering similar mobile users with respect to their habits.

/pdf/a-habit-mining-approach-for-discovering-similar-mobile-users-28mxn741cf.pdf

A habit mining approach for discovering similar mobile users

Multilayer perceptrons (MLPs) and support vector machines (SVMs) are flexible machine learning techniques that can fit complex nonlinear mappings. MLPs are the most popular neural network type, consisting on a feedforward network of processing neurons that are grouped into layers and connected by weighted links. On the other hand, SVM transforms the input variables into a high dimensional feature space and then finds the best hyperplane that models the data in the feature space. Both MLP and SVM are gaining an increase attention within the data mining (DM) field and are particularly useful when more simpler DM models fail to provide satisfactory predictive models. This tutorial chapter describes basic MLP and SVM concepts, under the CRISP-DM methodology, and shows how such learning tools can be applied to real-world classification and regression DM applications.

Data Mining: Foundations and Intelligent Paradigms

Behavioral Targeting (BT), which aims to deliver the most appropriate advertisements to the most appropriate users, is attracting much attention in online advertising market. A key challenge of BT is how to automatically segment users for ads delivery, and good user segmentation may significantly improve the ad click-through rate (CTR). Different from classical user segmentation strategies, which rarely take the semantics of user behaviors into consideration, we propose in this paper a novel user segmentation algorithm named Probabilistic Latent Semantic User Segmentation (PLSUS). PLSUS adopts the probabilistic latent semantic analysis to mine the relationship between users and their behaviors so as to segment users in a semantic manner. We perform experiments on the real world ad click through log of a commercial search engine. Comparing with the other two classical clustering algorithms, K-Means and CLUTO, PLSUS can further improve the ads CTR up to 100%. To our best knowledge, this work is an early semantic user segmentation study for BT in academia.

Probabilistic latent semantic user segmentation for behavioral targeted advertising

In this paper, we present an algorithm that can classify large-scale text data with high classification quality and fast training speed. Our method is based on a novel extension of the proximal SVM mode (Fung and Mangasarian, 2001). Previous studies on proximal SVM have focused on classification for low dimensional data and did not consider the unbalanced data cases. Such methods will meet difficulties when classifying unbalanced and high dimensional data sets such as text documents. In this work, we extend the original proximal SVM by learning a weight for each training error. We show that the classification algorithm based on this model is capable of handling high dimensional and unbalanced data. In the experiments, we compare our method with the original proximal SVM (as a special case of our algorithm) and the standard SVM (such as SVM light) on the recently published RCV1-v2 dataset. The results show that our proposed method had comparable classification quality with the standard SVM. At the same time, both the time and memory consumption of our method are less than that of the standard SVM.

/pdf/efficient-text-classification-by-weighted-proximal-svm-40yjm9mzl7.pdf

Efficient text classification by weighted proximal SVM

Understanding search intents of users through their condensed short queries has attracted much attention both in academia and industry. The search intents of users are generally assumed to be associated with various query patterns, such as "MobileName price", where "MobileName" could be any named entity of mobile phone model and this pattern indicates that the user intends to buy a mobile phone. However, discovering the query intent patterns for general search is challenging mainly due to the difficulty in collecting sufficient training data for learning query patterns across a large number of searchable domains. In this work, we propose Cross Domain Random Walk (CDRW) algorithm, which is semi-supervised, to discover the query intent patterns across different domains from search engine click-through log data. Starting with some manually tagged seed queries in one or more independent domains, CDRW takes the query patterns as bridge and propagates the transition probability across domains to collect the query intent patterns among different domains based on the assumption that "users who have similar intent in different but similar domains will have high probability to share similar query patterns across domains". Different from classical random walk algorithms, CDRW walks across different domains to disseminate the shared knowledge in a transfer learning manner. Extensive experiment results on real log data of a commercial search engine well validate the effectiveness and efficiency of the proposed algorithm.

Ying Chen

Papers

Probabilistic latent semantic user segmentation for behavioral targeted advertising

Efficient text classification by weighted proximal SVM

Cross Domain Random Walk for Query Intent Pattern Mining from Search Engine Log