Violetta Cavalli-Sforza

Transnational Information Sharing, Event Notification, Rule Enforcement and Process Coordination

Solutions to global problems such as disease detection and control, terrorism, immigration and border control, and illicit drug trafficking require sharing and coordinating information and collaboration among government agencies within a country and across national boundaries. This paper presents an approach to achieve information sharing, event notification, enforcement of policies, constraints, regulations, security and privacy rules, and process coordination. The proposed system, designed in collaboration with stakeholders and end users in two Latin American countries, achieves the desired capabilities by integrating a distributed query processor (DQP) that provides form-based and conversational user interfaces, a language translation system, an event server for event filtering and notification, and an event-trigger-rule server. The Web-services infrastructure is used to achieve the interoperation of these heterogeneous component systems. A prototype of the integrated transnational information system is described.

Bootstrapping a WordNet for an Arabic dialect from other WordNets and dictionary resources

We describe an experiment in developing a first version of WordNet for Iraqi Arabic starting from Arabic WordNet (for Modern Standard Arabic), Princeton WordNet (for English) and a bidirectional English-Iraqi Arabic dictionary. The resulting initial version of the target WordNet so-constructed was made available to human experts in Iraqi Arabic for correction and evaluation.

An Arabic Slot Grammar Parser

We describe a Slot Grammar (SG) parser for Arabic, ASG, and new features of SG designed to accommodate Arabic as well as the European languages for which SGs have been built. We focus on the integration of BAMA with ASG, and on a new, expressive SG grammar formalism, SGF, and illustrate how SGF is used to advantage in ASG.

Text readability for Arabic as a foreign language

In this study, we evaluate the informativeness of lexical, morphological and semantic features in determining the readability of texts geared towards learners of Arabic as a foreign language. We have gathered low-complexity features with the purpose of establishing a baseline for future research in readability assessment, using freely available natural language processing (NLP) and machine learning (ML) tools on a publicly accessible corpus. We tested common classification algorithms, as well as random forests-an ensemble learning method-and report on their results using several evaluation measures for comparability with similar work. Our results suggest that a small set of easily computed features can be indicative of the reading level of a text. Moreover, our findings will serve as a common ground, for ourselves and others, to evaluate and compare the performance of more elaborate techniques and feature sets.

Challenges in Adapting an Interlingua for Bidirectional English-Italian Translation

We describe our experience in adapting an existing high-quality, interlingual, unidirectional machine translation system to a new domain and bidirectional translation for a new language pair (English and Italian). We focus on the interlingua design changes which were necessary to achieve high quality output in view of the language mismatches between English and Italian. The representation we propose contains features that are interpreted differently, depending on the translation direction. This decision simplified the process of creating the interlingua for individual sentences, and allows the system to defer mapping of language-specific features (such as tense and aspect), which are realized when the target syntactic feature structure is created. We also describe a set of problems we encountered in translating modal verbs, and discuss the representation of modality in our interlingua.

AUI Story Maker: Animation Generation from Natural Language

Various research works have tried to connect Natural Language Processing NLP to computer graphics, as this connection would lay the ground for an automatic generation of computer animations. In this research we aim to provide a novel approach for connecting graphics to NLP by using OpenNLP and the Unity 3D game engine. We rely on two linguistic approaches—Vendler’s verb classification and Jackendoff’s Lexical Conceptual Structure LCS—and present how the technology enablers and the linguistic approaches chosen collaborate to provide the animation generation capability. We describe the overall architecture of AUI Story Maker, a system built to illustrate the feasibility of our approach, and discuss the future work required to make it a reliable tool in a modern classroom setting. We also present some writing samples gathered during field work with 1st graders at Al Akhawayn School in Ifrane (ASI), and provide sample outputs of AUI Story Maker.

Enhancing Visualization in Readability Reports for Arabic Texts.

Abstract Readability assessment for Arabic is still largely underserved in both research and software development. We believe that improved usability of the few tools currently released will motivate a greater user-base, and in doing so garner more interest in this topic from the research community. With that in mind, we examine recently developed readability tools with a graphical component, formulate recommendations, and propose visual enhancements to the way readability scores are reported to improve usability and informativeness.

Arabic Computational Morphology: A Trade-off Between Multiple Operations and Multiple Stems

We present a computational approach to Arabic morphology description that draws from Lexeme-Based Morphology (Aronoff, 1994; Beard, 1995), giving priority to stems and granting a subordinate status to inflectional prefixes and suffixes. Although the morphology of Arabic is non-concatenative, we make the process of generating inflected forms concatenative by separating the generation of stems from that of other inflectional affixes. Our approach is implemented in an extension of the MORPHĒ tool (Leavitt, 1994), which has been enhanced in order to provide a representational formalism that embodies Lexeme-Based Morphology theory and minimizes the number of rules required for the description of Arabic morphology

Intelligent Learning by Doing Tools for Technical and Dialectical Knowledge

New members entering productive organizations require considerable training. Computer tools can support such training by providing an opportunity to learn while engaging in authentic activities and receiving appropriate coaching. We describe two tools that incorporate this approach. Sherlock, an existing computer coach, is an effective environment for learning how to troubleshoot complex electronic devices. A newer research effort focuses on tools for supporting knowledge-building argumentation and scientific theory evaluation in post-elementary school science education. Both tools offer users opportunities for reflecting on their own performance and support individual as well as collaborative learning.