Abraham Bagherjeiran

Combining Behavioral and Social Network Data for Online Advertising

There are two main requirements for effective advertising in social networks. The first is that links in the social network are relevant to the targeted ads. The second is that social information can be easily incorporated with existing targeting methods to predict response rates. Our purpose in this paper is to investigate these requirements. We measure the relevance of a social network, the Yahoo! Instant Messenger graph, to classes of ads. We investigate the degree to which social network information complements existing user-profile information for targeting. We find that there is significant evidence in our social network of homophily, that links in the network indicate similar ad-relevant interests. We propose an ensemble classifier to combine existing user-only models with social network features to improve response predictions.

Using clustering to learn distance functions for supervised similarity assessment

Assessing the similarity between objects is a prerequisite for many data mining techniques. This paper introduces a novel approach to learn distance functions that maximizes the clustering of objects belonging to the same class. Objects belonging to a data set are clustered with respect to a given distance function and the local class density information of each cluster is then used by a weight adjustment heuristic to modify the distance function so that the class density is increased in the attribute space. This process of interleaving clustering with distance function modification is repeated until a “good” distance function has been found. We implemented our approach using the k-means clustering algorithm. We evaluated our approach using 7 UCI data sets for a traditional 1-nearest-neighbor (1-NN) classifier and a compressed 1-NN classifier, called NCC, that uses the learnt distance function and cluster centroids instead of all the points of a training set. The experimental results show that attribute weighting leads to statistically significant improvements in prediction accuracy over a traditional 1-NN classifier for 2 of the 7 data sets tested, whereas using NCC significantly improves the accuracy of the 1-NN classifier for 4 of the 7 data sets.

Adaptive clustering: obtaining better clusters using feedback and past experience

Adaptive clustering uses external feedback to improve cluster quality; past experience serves to speed up execution time. An adaptive clustering environment is proposed that uses Q-learning to learn the reward values of successive data clusterings. Adaptive clustering supports the reuse of clusterings by memorizing what worked well in the past. It has the capability of exploring multiple paths in parallel when searching for good clusters. In a case study, we apply adaptive clustering to instance-based learning relying on a distance function modification approach. A distance function adaptation scheme that uses external feedback is proposed and compared with other distance function learning approaches. Experimental results indicate that the use of adaptive clustering leads to significant improvements of instance-based learning techniques, such as k-nearest neighbor classifiers. Moreover, as a by-product a new instance-based learning technique is introduced that classifies examples by solely using cluster representatives; this technique shows high promise in our experimental evaluation.

Distance Function Learning for Supervised Similarity Assessment

Assessing the similarity between cases is a prerequisite for many case-based reasoning tasks. This chapter centers on distance function learning for supervised similarity assessment. First a framework for supervised similarity assessment is introduced. Second, three supervised distance function learning approaches from the areas of pattern classification, supervised clustering, and information retrieval are discussed, and their results for two supervised learning tasks will be explained and visualized. In each of these different areas, we show how the method can be applied to areas of case-based reasoning. Finally, a detailed literature survey will be given.

Online Advertising in Social Networks

Online social networks offer opportunities to analyze user behavior and social connectivity and leverage resulting insights for effective online advertising. This chapter focuses on the role of social network information in online display advertising.

Content-based image retrieval through a multi-agent meta-learning framework

The objective of a general-purpose content-based image retrieval system is to find images in a database that match an external measure of relevance. Since users follow different and inconsistent relevance measures, processing queries in a task-specific manner has shown to be an effective approach. Viewing specialized image retrieval algorithms as agents, we propose a general-purpose image retrieval system that uses a new multi-agent meta-learning framework. The framework adapts a distance function defined over both image distance weights and image queries to identify clusters of algorithms that produce similar solutions to similar problems. Experiments compare our approach with a traditional information retrieval algorithm; results show that our framework provides better average relevance scores

A framework for spatial feature selection and scoping and its application to geo-targeting

Predicting if a particular user clicks on a particular ad is of critical importance for internet advertising. Associations between Internet ad performance data, such as number of clicks or Click Through Rate, CTR, and demographic data may be very weak on the global level, but strong at the regional level. Identifying regions with strong associations of a continuous performance attribute with geo-features can create valuable knowledge for geo-targeted advertising. In this paper, we present a novel framework for interestingness scoping to identify such regions and discuss how such interestingness hotspots can be used for geo-feature evaluation with the goal to develop more accurate prediction models for advertisers. We also present the ZIPS algorithm that takes initial seed zip codes and discovers interestingness hotspots/coldspots, and a geo-feature preselection algorithm which automatically finds promising geo-features and identifies initial seed zipcodes for the ZIPS algorithm. We applied our framework to a large number of geo-spatial data sets, combining data from a major ad network, demographic data from the 2000 Census, and binary feature data from other sources. Our experimental results demonstrate that creating geo-features can double CTR performance for an Ad.