We study astroturf political campaigns on microblogging platforms: politically-motivated individuals and organizations that use multiple centrally-controlled accounts to create the appearance of widespread support for a candidate or opinion. We describe a machine learning framework that combines topological, content-based and crowdsourced features of information diffusion networks on Twitter to detect the early stages of viral spreading of political misinformation.  We present promising preliminary results with better than 96% accuracy in the detection of astroturf content in the run-up to the 2010 U.S. midterm elections.

Online discussion boards, or Internet forums, are a significant part of the Internet. People use Internet forums to post questions, provide advice and participate in discussions. These online conversations are represented as threads, and the conversation trees within these threads are important in understanding the behaviour of online users. Unfortunately, the reply structures of these threads are generally not publicly accessible or not maintained. Hence, in this paper, we introduce an efficient and simple approach to reconstruct the reply structure in threaded conversations. We contrast its accuracy against three baseline algorithms, and show that our algorithm can accurately recreate the in and out degree distributions of forum reply graphs built from the reconstructed reply structures.

Viral marketing mechanisms use the existing social network between customers to spread information about products and encourage product adoption. Existing viral marketing models focus on the dynamics of the diffusion process, however they typically: (a) only consider a single product campaign and (b) fail to model the evolution of the social network, as the trust between individuals changes over time, during the course of multiple campaigns. In this work, we propose an adaptive viral marketing model which captures: (1) multiple different product campaigns, (2) the diversity in customer preferences among different product categories, and (3) changing confidence in peers’ recommendations over time. By applying our model to a real-world network extracted from the Digg social news website, we provide insights into the effects of network dynamics on the different products’ adoption. Our experiments show that our proposed model outperforms earlier nonadaptive diffusion models in predicting future product adoptions. We also show how this model can be used to explore new viral marketing strategies that are more successful than classic strategies which ignore the dynamic nature of social networks.

In this article, we present a method for predicting the view count of a YouTube video using a small feature set collected from a synchronous sharing tool. We hypothesize that videos which have a high YouTube view count will exhibit a unique sharing pattern when shared in synchronous environments. Using a one-day sample of 2,188 dyadic sessions from the Yahoo! Zync synchronous sharing tool, we demonstrate how to predict the video's view count on YouTube, specifically if a video has over 10 million views. The prediction model is 95.8% accurate and done with a relatively small training set; only 15% of the videos had more than one session viewing; in effect, the classifier had a precision of 76.4% and a recall of 81%. We describe a prediction model that relies on using implicit social shared viewing behavior such as how many times a video was paused, rewound, or fast-forwarded as well as the duration of the session. Finally, we present some new directions for future virality research and for the design of future social media tools.

Software defect prediction aims to reduce software testing efforts by guiding testers through the defect-prone sections of software systems. Defect predictors are widely used in organizations to predict defects in order to save time and effort as an alternative to other techniques such as manual code reviews. The usage of a defect prediction model in a real-life setting is difficult because it requires software metrics and defect data from past projects to predict the defect-proneness of new projects. It is, on the other hand, very practical because it is easy to apply, can detect defects using less time and reduces the testing effort. We have built a learning-based defect prediction model for a telecommunication company in the space of one year. In this study, we have briefly explained our model, presented its pay-off and described how we have implemented the model in the company. Furthermore, we compared the performance of our model with that of another testing strategy applied in a pilot project that implemented a new process called Team Software Process (TSP). Our results show that defect predictors can predict 87 percent of code defects, decrease inspection efforts by 72 percent and hence, reduces post-release defects by 44 percent. Furthermore, they can be used as complementary tools for a new process implementation whose effects on testing activities are limited.

Transfer learning aims to solve new learning problems by extracting and making use of the common knowledge found in related domains. A key element of transfer learning is to identify structured knowledge to enable the knowledge transfer. Structured knowledge comes in different forms, depending on the nature of the learning problem and characteristics of the domains. In this article, we describe three of our recent works on transfer learning in a progressively more sophisticated order of the structured knowledge being transferred. We show that optimization methods, and techniques inspired by the concerns of data reuse can be applied to extract and transfer deep structural knowledge between a variety of source and target problems. In our examples, this knowledge spans explicit data labels, model parameters, relations between data clusters and relational action descriptions. 

We propose a model for errors in sung queries, a variant of the hidden Markov model (HMM). This is a solution to the problem of identifying the degree of similarity between a (typically error-laden) sung query and a potential target in a database of musical works, an important problem in the field of music information retrieval. Similarity metrics are a critical component of query-by-humming (QBH) applications which search audio and multimedia databases for strong matches to oral queries. Our model comprehensively expresses the types of error or variation between target and query: cumulative and non-cumulative local errors, transposition, tempo and tempo changes, insertions, deletions and modulation. The model is not only expressive, but automatically trainable, or able to learn and generalize from query examples. We present results of simulations, designed to assess the discriminatory potential of the model, and tests with real sung queries, to demonstrate relevance to real-world applications.

Motivation: Biomarker discovery from high-dimensional data is a crucial problem with enormous applications in biology and medicine. It is also extremely challenging from a statistical viewpoint, but surprisingly few studies have investigated the relative strengths and weaknesses of the plethora of existing feature selection methods. Methods: We compare 32 feature selection methods on 4 public gene expression datasets for breast cancer prognosis, in terms of predictive performance, stability and functional interpretability of the signatures they produce. Results: We observe that the feature selection method has a significant influence on the accuracy, stability and interpretability of signatures. Simple filter methods generally outperform more complex embedded or wrapper methods, and ensemble feature selection has generally no positive effect. Overall a simple Student's t-test seems to provide the best results. Availability: Code and data are publicly available at http://cbio.ensmp.fr/~ahaury/.

The proliferation of online information sources has led to an increased use of wrappers for extracting data from Web sources. While most of the previous research has focused on quick and efficient generation of wrappers, the development of tools for wrapper maintenance has received less attention. This is an important research problem because Web sources often change in ways that prevent the wrappers from extracting data correctly. We present an efficient algorithm that learns structural information about data from positive examples alone. We describe how this information can be used for two wrapper maintenance applications: wrapper verification and reinduction. The wrapper verification system detects when a wrapper is not extracting correct data, usually because the Web source has changed its format. The reinduction algorithm automatically recovers from changes in the Web source by identifying data on Web pages so that a new wrapper may be generated for this source. To validate our approach, we monitored 27 wrappers over a period of a year. The verification algorithm correctly discovered 35 of the 37 wrapper changes, and made 16 mistakes, resulting in precision of 0.73 and recall of 0.95. We validated the reinduction algorithm on ten Web sources. We were able to successfully reinduce the wrappers, obtaining precision and recall values of 0.90 and 0.80 on the data extraction task.

For large, real-world inductive learning problems, the number of training examples often must be limited due to the costs associated with procuring, preparing, and storing the training examples and/or the computational costs associated with learning from them. In such circumstances, one question of practical importance is: if only n training examples can be selected, in what proportion should the classes be represented? In this article we help to answer this question by analyzing, for a fixed training-set size, the relationship between the class distribution of the training data and the performance of classification trees induced from these data. We study twenty-six data sets and, for each, determine the best class distribution for learning. The naturally occurring class distribution is shown to generally perform well when classifier performance is evaluated using undifferentiated error rate (0/1 loss). However, when the area under the ROC curve is used to evaluate classifier performance, a balanced distribution is shown to perform well. Since neither of these choices for class distribution always generates the best-performing classifier, we introduce a budget-sensitive progressive sampling algorithm for selecting training examples based on the class associated with each example. An empirical analysis of this algorithm shows that the class distribution of the resulting training set yields classifiers with good (nearly-optimal) classification performance.