An increasing amount of data is published as semistructured documents formatted with presentational markup. Examples include data objects such as mathematical expressions encoded with MathML or web pages encoded with XHTML. Our intention is to improve the state of the art in retrieving, manipulating, or mining such data. We focus first on mathematics retrieval, which is appealing in various domains, such as education, digital libraries, engineering, patent documents, and medical sciences. Capturing the similarity of mathematical expressions also greatly enhances document classification in such domains. Unlike text retrieval, where keywords carry enough semantics to distinguish text documents and rank them, math symbols do not contain much semantic information on their own. Unfortunately, considering the structure of mathematical expressions to calculate relevance scores of documents results in ranking algorithms that are computationally more expensive than the typical ranking algorithms employed for text documents. As a result, current math retrieval systems either limit themselves to exact matches, or they ignore the structure completely; they sacrifice either recall or precision for efficiency. We propose instead an efficient end-to-end math retrieval system based on a structural similarity ranking algorithm. We describe novel optimization techniques to reduce the index size and the query processing time. Thus, with the proposed optimizations, mathematical contents can be fully exploited to rank documents in response to mathematical queries. We demonstrate the effectiveness and the efficiency of our solution experimentally, using a special-purpose testbed that we developed for evaluating math retrieval systems. We finally extend our retrieval system to accommodate rich queries that consist of combinations of math expressions and textual keywords. As a second focal point, we address the problem of recognizing structural repetitions in typical web documents. Most web pages use presentational markup standards, in which the tags control the formatting of documents rather than semantically describing their contents. Hence, their structures typically contain more irregularities than descriptive (data-oriented) markup languages. Even though applications would greatly benefit from a grammar inference algorithm that captures structure to make it explicit, the existing algorithms for XML schema inference, which target data-oriented markup, are ineffective in inferring grammars for web documents with presentational markup. There is currently no general-purpose grammar inference framework that can handle irregularities commonly found in web documents and that can operate with only a few examples. Although inferring grammars for individual web pages has been partially addressed by data extraction tools, the existing solutions rely on simplifying assumptions that limit their application. Hence, we describe a principled approach to the problem by defining a class of grammars that can be inferred from very small sample sets and can capture the structure of most web documents. The effectiveness of this approach, together with a comparison against various classes of grammars including DTDs and XSDs, is demonstrated through extensive experiments on web documents. We finally use the proposed grammar inference framework to extend our math retrieval system and to optimize it further.

An increasing amount of data is published as semistructured documents formatted with presentational markup. Examples include data objects such as mathematical expressions encoded with MathML or web pages encoded with XHTML. Our intention is to improve the state of the art in retrieving, manipulating, or mining such data. We focus first on mathematics retrieval, which is appealing in various domains, such as education, digital libraries, engineering, patent documents, and medical sciences. Capturing the similarity of mathematical expressions also greatly enhances document classification in such domains. Unlike text retrieval, where keywords carry enough semantics to distinguish text documents and rank them, math symbols do not contain much semantic information on their own. Unfortunately, considering the structure of mathematical expressions to calculate relevance scores of documents results in ranking algorithms that are computationally more expensive than the typical ranking algorithms employed for text documents. As a result, current math retrieval systems either limit themselves to exact matches, or they ignore the structure completely; they sacrifice either recall or precision for efficiency. We propose instead an efficient end-to-end math retrieval system based on a structural similarity ranking algorithm. We describe novel optimization techniques to reduce the index size and the query processing time. Thus, with the proposed optimizations, mathematical contents can be fully exploited to rank documents in response to mathematical queries. We demonstrate the effectiveness and the efficiency of our solution experimentally, using a special-purpose testbed that we developed for evaluating math retrieval systems. We finally extend our retrieval system to accommodate rich queries that consist of combinations of math expressions and textual keywords. As a second focal point, we address the problem of recognizing structural repetitions in typical web documents. Most web pages use presentational markup standards, in which the tags control the formatting of documents rather than semantically describing their contents. Hence, their structures typically contain more irregularities than descriptive (data-oriented) markup languages. Even though applications would greatly benefit from a grammar inference algorithm that captures structure to make it explicit, the existing algorithms for XML schema inference, which target data-oriented markup, are ineffective in inferring grammars for web documents with presentational markup. There is currently no general-purpose grammar inference framework that can handle irregularities commonly found in web documents and that can operate with only a few examples. Although inferring grammars for individual web pages has been partially addressed by data extraction tools, the existing solutions rely on simplifying assumptions that limit their application. Hence, we describe a principled approach to the problem by defining a class of grammars that can be inferred from very small sample sets and can capture the structure of most web documents. The effectiveness of this approach, together with a comparison against various classes of grammars including DTDs and XSDs, is demonstrated through extensive experiments on web documents. We finally use the proposed grammar inference framework to extend our math retrieval system and to optimize it further.

In the information age of today, vast amounts of sensitive and confidential data is exchanged over an array of different mediums. Accompanied with this phenomenon is a comparable increase in the number and types of attacks to acquire this information. Information security and data consistency have hence, become quintessentially important. Log file analysis has proven to be a good defense mechanism as logs provide an accessible record of network activities in the form of server generated messages. However, manual analysis is tedious and prohibitively time consuming. Traditional log analysis techniques, based on pattern matching and data mining approaches, are ad hoc and cannot readily adapt to different kinds of log files. The goal of this research is to explore the use of grammar inference for log file analysis in order to build a more adaptive, flexible and generic method for message categorization, anomaly detection and reporting. The grammar inference process employs robust parsing, islands grammars and source transformation techniques. We test the system by using three different kinds of log file training sets as input and infer a grammar and generate message categories for each set. We detect anomalous messages in new log files using the inferred grammar as a catalog of valid traces and present a reporting program to extract the instances of specified message categories from the log files. / Thesis (Master, Computing) -- Queen's University, 2008-05-22 14:12:30.199

Die automatische Informationsextraktion (IE) aus unstrukturierten Texten ermöglicht völlig neue Wege, auf relevante Informationen zuzugreifen und deren Inhalte zu analysieren, die weit über bisherige Verfahren zur Stichwort-basierten Dokumentsuche hinausgehen. Die Entwicklung von Programmen zur Extraktion von maschinenlesbaren Daten aus Texten erfordert jedoch nach wie vor die Entwicklung von domänenspezifischen Extraktionsprogrammen. Insbesondere im Bereich der Enterprise Search (der Informationssuche im Unternehmensumfeld), in dem eine große Menge von heterogenen Dokumenttypen existiert, ist es oft notwendig ad-hoc Programm-module zur Extraktion von geschäftsrelevanten Entitäten zu entwickeln, die mit generischen Modulen in monolithischen IE-Systemen kombiniert werden. Dieser Umstand ist insbesondere kritisch, da potentiell für jeden einzelnen Anwendungsfall ein von Grund auf neues IE-System entwickelt werden muss. Die vorliegende Dissertation untersucht die effiziente Entwicklung und Ausführung von IE-Systemen im Kontext der Enterprise Search und effektive Methoden zur Ausnutzung bekannter strukturierter Daten im Unternehmenskontext für die Extraktion und Identifikation von geschäftsrelevanten Entitäten in Doku-menten. Grundlage der Arbeit ist eine neuartige Plattform zur Komposition von IE-Systemen auf Basis der Beschreibung des Datenflusses zwischen generischen und anwendungsspezifischen IE-Modulen. Die Plattform unterstützt insbesondere die Entwicklung und Wiederverwendung von generischen IE-Modulen und zeichnet sich durch eine höhere Flexibilität und Ausdrucksmächtigkeit im Vergleich zu vorherigen Methoden aus. Ein in der Dissertation entwickeltes Verfahren zur Dokumentverarbeitung interpretiert den Daten-austausch zwischen IE-Modulen als Datenströme und ermöglicht damit eine weitgehende Parallelisierung von einzelnen Modulen. Die autonome Ausführung der Module führt zu einer wesentlichen Beschleu-nigung der Verarbeitung von Einzeldokumenten und verbesserten Antwortzeiten, z. B. für Extraktions-dienste. Bisherige Ansätze untersuchen lediglich die Steigerung des durchschnittlichen Dokumenten-durchsatzes durch verteilte Ausführung von Instanzen eines IE-Systems. Die Informationsextraktion im Kontext der Enterprise Search unterscheidet sich z. B. von der Extraktion aus dem World Wide Web dadurch, dass in der Regel strukturierte Referenzdaten z. B. in Form von Unternehmensdatenbanken oder Terminologien zur Verfügung stehen, die oft auch die Beziehungen von Entitäten beschreiben. Entitäten im Unternehmensumfeld haben weiterhin bestimmte Charakteristiken: Eine Klasse von relevanten Entitäten folgt bestimmten Bildungsvorschriften, die nicht immer bekannt sind, auf die aber mit Hilfe von bekannten Beispielentitäten geschlossen werden kann, so dass unbekannte Entitäten extrahiert werden können. Die Bezeichner der anderen Klasse von Entitäten haben eher umschreibenden Charakter. Die korrespondierenden Umschreibungen in Texten können variieren, wodurch eine Identifikation derartiger Entitäten oft erschwert wird. Zur effizienteren Entwicklung von IE-Systemen wird in der Dissertation ein Verfahren untersucht, das alleine anhand von Beispielentitäten effektive Reguläre Ausdrücke zur Extraktion von unbekannten Entitäten erlernt und damit den manuellen Aufwand in derartigen Anwendungsfällen minimiert. Verschiedene Generalisierungs- und Spezialisierungsheuristiken erkennen Muster auf verschiedenen Abstraktionsebenen und schaffen dadurch einen Ausgleich zwischen Genauigkeit und Vollständigkeit bei der Extraktion. Bekannte Regellernverfahren im Bereich der Informationsextraktion unterstützen die beschriebenen Problemstellungen nicht, sondern benötigen einen (annotierten) Dokumentenkorpus. Eine Methode zur Identifikation von Entitäten, die durch Graph-strukturierte Referenzdaten vordefiniert sind, wird als dritter Schwerpunkt untersucht. Es werden Verfahren konzipiert, welche über einen exakten Zeichenkettenvergleich zwischen Text und Referenzdatensatz hinausgehen und Teilübereinstimmungen und Beziehungen zwischen Entitäten zur Identifikation und Disambiguierung heranziehen. Das in der Arbeit vorgestellte Verfahren ist bisherigen Ansätzen hinsichtlich der Genauigkeit und Vollständigkeit bei der Identifikation überlegen. / The automatic information extraction (IE) from unstructured texts enables new ways to access relevant information and analyze text contents, which goes beyond existing technologies for keyword-based search in document collections. However, the development of systems for extracting machine-readable data from text still requires the implementation of domain-specific extraction programs. In particular in the field of enterprise search (the retrieval of information in the enterprise settings), in which a large amount of heterogeneous document types exists, it is often necessary to develop ad-hoc program-modules and to combine them with generic program components to extract by business relevant entities. This is particularly critical, as potentially for each individual application a new IE system must be developed from scratch. In this work we examine efficient methods to develop and execute IE systems in the context of enterprise search and effective algorithms to exploit pre-existing structured data in the business context for the extraction and identification of business entities in documents. The basis of this work is a novel platform for composition of IE systems through the description of the data flow between generic and application-specific IE modules. The platform supports in particular the development and reuse of generic IE modules and is characterized by a higher flexibility as compared to previous methods. A technique developed in this work interprets the document processing as data stream between IE modules and thus enables an extensive parallelization of individual modules. The autonomous execution of each module allows for a significant runtime improvement for individual documents and thus improves response times, e.g. for extraction services. Previous parallelization approaches focused only on an improved throughput for large document collections, e.g., by leveraging distributed instances of an IE system. Information extraction in the context of enterprise search differs for instance from the extraction from the World Wide Web by the fact that usually a variety of structured reference data (corporate databases or terminologies) is available, which often describes the relationships among entities. Furthermore, entity names in a business environment usually follow special characteristics: On the one hand relevant entities such as product identifiers follow certain patterns that are not always known beforehand, but can be inferred using known sample entities, so that unknown entities can be extracted. On the other hand many designators have a more descriptive character (concatenation of descriptive words). The respective references in texts might differ due to the diversity of potential descriptions, often making the identification of such entities difficult. To address IE applications in the presence of available structured data, we study in this work the inference of effective regular expressions from given sample entities. Various generalization and specialization heuristics are used to identify patterns at different syntactic abstraction levels and thus generate regular expressions which promise both high recall and precision. Compared to previous rule learning techniques in the field of information extraction, our technique does not require any annotated document corpus. A method for the identification of entities that are predefined by graph structured reference data is examined as a third contribution. An algorithm is presented which goes beyond an exact string comparison between text and reference data set. It allows for an effective identification and disambiguation of potentially discovered entities by exploitation of approximate matching strategies. The method leverages further relationships among entities for identification and disambiguation. The method presented in this work is superior to previous approaches with regard to precision and recall.