This thesis focuses on Arabic embedded text detection and recognition in videos. Different approaches robust to Arabic text variability (fonts, scales, sizes, etc.) as well as to environmental and acquisition condition challenges (contrasts, degradation, complex background, etc.) are proposed. We introduce different machine learning-based solutions for robust text detection without relying on a...

GO/Max is an agent programming language that facilitates the design of algorithms for real-time control of sound/music generation programs crafted in the Max/MSP environment. We show how software planning agents programmed in GO/Max can be used to transform abstract goal states specified by the performer in potentially complex sequences of Max/MSP control messages.

This paper extends standard help system technology to demonstrate the suitability of Formal Concept Analysis in displaying, searching and navigating help content. The paper introduces a method for building suitable scales directly from the help system index by computing a keyword extension set. The keyword extension technique is generalisable in any document collection where a hand-crafted inde...

We propose a data-driven method for automatically analyzing the morphology of ancient Greek. This method improves on existing ancient Greek analyzers in two ways. First, through the use of a nearestneighbor machine learning framework, the analyzer requires no hand-crafted rules. Second, it is able to predict novel roots, and to rerank its predictions by exploiting a large, unlabelled corpus of ...

We present a solution for the Linked Data Mining Challenge 2016, that achieved 92.5% accuracy according to the submission system. The solution uses a hand-crafted dataset, that was created by scraping various websites for reviews. We use logistic regression to learn a classification model and we publish all our results to GitHub.

Options have been shown to be a key step in extending reinforcement learning beyond low-level reactionary systems to higher-level, planning systems. Most of the options research involves hand-crafted options; there has been only very limited work in the automated discovery of options. We extend early work in automated option discovery with a flexible and robust method.

The Vertica SQL Query Optimizer was written from the ground up for the Vertica Analytic Database. Its design, and the tradeoffs we encountered during implementation, support the case that the full power of novel database systems can be realized only with a custom Query Optimizer, carefully crafted exclusively for the system in which it operates.

We introduce an annotation scheme for temporal expressions, and describe a method for resolving temporal expressions in print and broadcast news. The system, which is based on both hand-crafted and machine-learnt rules, achieves an 83.2% accuracy (Fmeasure) against hand-annotated data. Some initial steps towards tagging event chronologies are also described.

In this short note we present a collection of benchmarks submitted to the SAT Competition 2016. Most of them stem from other sources, some crafted ones are new, but all present equivalence checking problems (miters) for arithmetic circuits, such as multipliers.

In this work we present the solution to the DEBS’2013 Grand Challenge, as crafted by the joint effort of teams from the Technion and TU Dortmund. The paper describes the architecture, details the queries and offers our observations regarding the appropriate way to trade-off high-level processing with time constraints.