Abstract
In soccer, the level of performance is determined by a number of a complex variables interrelated: technique, tactics, psychological factors and finally, fitness. Because of this, analyzing this information in a real-time, even for soccer experts like professional coaches has become an impossible task. Automatic event detection tools occupy an important role in this reality, although nowadays there isn't any tool capable of producing information capable of helping a professional coach choosing his team strategy for a specific game. In this research project an automatic event detection tool is purposed and, a set of game statistics defined by a group of sports researchers. All the teams present in the 2009 RoboCup tournament have a pass success rate superior to 65%. These statistics provide an interesting viewpoint on how to evaluate a team performance, such as the importance of dominating the opposing team field without losing the control of our own (this can be seen on the top 3 zone dominance statistics). In the future this project will serve as a base for building a Framework capable of simulating a match between two heterogeneous soccer teams and produce reliable information for optimizing the team performance.

Abstract
The problem of joint modeling the text and image components of multimedia documents is studied. The text component is represented as a sample from a hidden topic model, learned with latent Dirichlet allocation, and images are represented as bags of visual (SIFT) features. Two hypotheses are investigated: that 1) there is a benefit to explicitly modeling correlations between the two components, and 2) this modeling is more effective in feature spaces with higher levels of abstraction. Correlations between the two components are learned with canonical correlation analysis. Abstraction is achieved by representing text and images at a more general, semantic level. The two hypotheses are studied in the context of the task of cross-modal document retrieval. This includes retrieving the text that most closely matches a query image, or retrieving the images that most closely match a query text. It is shown that accounting for cross-modal correlations and semantic abstraction both improve retrieval accuracy. The cross-modal model is also shown to outperform state-of-the-art image retrieval systems on a unimodal retrieval task. © 2010 ACM.

Abstract
In order to fulfill all challenges related to design, management, evaluation and reconfiguration of new or existing facilities, the development of a new, integrated and holistic factory framework is required. In this context, a new requirements model is proposed in the context of the innovative framework called Virtual Factory Framework. Using this approach in factory planning it is expected to improve the ability to generate better and more sustainable solutions over the entire factory life-cycle. This paper underlines the use of the life-cycle paradigm in order to enable the future factory planning approaches.

Abstract
Meiotic and early-embryonic cell divisions in vertebrates take place in the absence of transcription and rely on the translational regulation of stored maternal messenger RNAs. Most of these mRNAs are regulated by the cytoplasmic-polyadenylation-element-binding protein (CPEB), which mediates translational activation and repression through cytoplasmic changes in their poly(A) tail length. It was unknown whether translational regulation by cytoplasmic polyadenylation and CPEB can also regulate mRNAs at specific points of mitotic cell-cycle divisions. Here we show that CPEB-mediated post-transcriptional regulation by phase-specific changes in poly(A) tail length is required for cell proliferation and specifically for entry into M phase in mitotically dividing cells. This translational control is mediated by two members of the CPEB family of proteins, CPEB1 and CPEB4. We conclude that regulation of poly(A) tail length is not only required to compensate for the lack of transcription in specialized cell divisions but also acts as a general mechanism to control mitosis.

Abstract
Breast Cancer Conservative Treatment (BCCT) is considered nowadays to be the most widespread form of locor-regional breast cancer treatment. However, aesthetic results are heterogeneous and difficult to evaluate in a standardized way. The limited reproducibility of subjective aesthetic evaluation in BCCT motivated the research towards objective methods. A recent computer system (BCCT. core) was developed to objectively and automatically evaluate the aesthetic result of BCCT. The system is centered on a support vector machine (SVM) classifier with a radial basis function (RBF) used to predict the overall cosmetic result from features computed on a digital photograph of the patient. However, this classifier is not ideal for the interpretation of the factors being used in the prediction. Therefore, an often suggested improvement is the interpretability of the model being used to assess the overall aesthetic result. In the current work we investigate the accuracy of different interpretable methods against the model currently deployed in the BCCT. core software. We compare the performance of decision trees and linear classifiers with the RBF SVM currently in BCCT. core. In the experimental study, these interpretable models shown a similar accuracy to the currently used RBF SVM, suggesting that the later can be replaced without sacrificing the performance of the BCCT.core.

Abstract