Information Extraction Rules Research Articles

A knowledge-based intelligent recognition method for rock discontinuities with point cloud data

ABSTRACT Rock discontinuity information plays a key role in engineering activities. In severely fragmented and discontinuous samples, the identification accuracy of rock discontinuities can be improved with the help of geologic knowledge with identified spatial features and geoscience interaction mechanisms. Therefore, we proposed a new method – a knowledge-based intelligent recognition method for rock discontinuities that used point cloud data – to correct inaccurate identification by considering the spatial relationships of rock discontinuities. This paper integrated the geologic knowledge bases in rock discontinuity information extraction rules and combined the generative rule inference model with the region growing method for intelligent extraction. In the case study, we developed four typical rule sets as examples based on the constructed rule system to group rock discontinuities and subsume the fractured rock discontinuities. Results show that this new method is effective in identifying and classifying rock discontinuities of complex areas with a greater reduction in manual intervention.

Design of social media information extraction system based on deep learning

Aiming at the problems of low accuracy and long time in traditional systems, a social media information extraction system based on deep learning is designed. Firstly, the overall framework of the system is designed, including text extraction module, keyword extraction module and emotion analysis module. Then, the social media information is preprocessed, the emotional resource establishment and information extraction rules are constructed according to the preprocessing results, and the convolution neural network is used to construct the social media information extraction model. Finally, according to the correlation between text entries and categories, the global MI values of entries and all categories are calculated. The calculation results are inputted into the constructed convolution neural network model, and the social media information extraction results are output. The simulation results show that the extraction accuracy of the designed system is high and the extraction time is within 15 s.

A deep neural network-based method for deep information extraction using transfer learning strategies to support automated compliance checking

Existing automated compliance checking (ACC) systems require the extraction of requirements from regulatory documents into computer-processable representations. These information extraction (IE) processes are either fully manual, semi-automated, or automated. Semi-automated and manual approaches typically use manual annotations or predefined IE rules, which lack sufficient flexibility and scalability; the annotations and rules typically need adaptation if the characteristics of the regulatory document change. There is, thus, a need for a fully automated IE approach that can achieve high and consistent performance across different types of regulatory documents for supporting ACC. To address this need, this paper proposes a deep neural network-based method for deep IE – extracting semantic and syntactic information elements – from regulatory documents in the architectural, engineering, and construction (AEC) domain. The proposed method was evaluated in extracting information from multiple regulatory documents in the AEC domain. It achieved average precision and recall of 93.1% and 92.9%, respectively.

A Heuristic Approach for Telugu Text Summarization with Improved Sentence Ranking

Extracting/abstracting the condensed form of original text document by retaining its information and complete meaning is known as text summarization. The creation of manual summaries from large text documents is difficult and time-consuming for humans. Text summarization has become an important and challenging area in natural language processing. This paper presents a heuristic appraoch to extract a summary of e-news articles of the Telugu language. The method proposes new lexical parameter-based information extraction (IE) rules for scoring the sentences. Event score and Named Entity Score is a novel part in sentence scoring to identify the essential information in the text. Depending on the frequency of occurrence of event/named entites in the sentence and document, sentences are selected for summary. Data is collected from online news sources (i.e., Eenadu, Sakshi,Andhra Jyothi, Namaste Telangana) to experiment. The proposed method is compared with other techniques developed for Telugu text summarization. Evaluation metrics like precision, recall, and F1 score is used to measure the proposed method's performance. An extensive statistical and qualitative evaluation of the system's summaries has been conducted using Recall-Oriented Understudy for Gisting Evaluation (ROUGE), a standard summary evaluation tool. The results showed improved performance compared to other methods.

Technical Perspective

The challenge of extracting structured information from text, or sequential data in general, is prevalent across a multitude of data-science domains. This challenge, known as Information Extraction (IE), instantiates to core components in text analytics, and a plethora of IE paradigms have been developed over the past decades. Rules and rule systems have consistently been key components in such paradigms, yet their roles have varied and evolved over time. Analytics engines such as IBM's SystemT use IE rules for materializing relations inside relational query languages. Machinelearning classifiers and probabilistic graphical models (e.g., Conditional Random Fields) use rules for feature generation. They also serve as weak constraints in Markov Logic Networks (and extensions such as DeepDive), and generators of noisy training data in the state-of-the-art Snorkel system.

Abs-Sum-Kan: An Abstractive Text Summarization Technique for an Indian Regional Language by Induction of Tagging Rules

This paper presents a full abstraction for Indian languages, specifically Kannada, in the context of guided summarization. The proposed process generates the abstractive summary by focusing on a unified presentation model with aspect based Information Extraction (IE) rules and scheme based Templates. TF/IDF rules are used for classification into categories. Lexical analysis (like Parts Of Speech tagging and Named Entity Recognition) reduces prolixity, which leads to robust IE rules. Usage of Templates for sentence generation makes the summaries succinct and information intensive. The IE rules are designed to accommodate the complexities of the considered languages. Later, the system aims to produce a guided summary of domain specific documents. An abstraction scheme is a collection of aspects and associated IE rules. Each abstraction scheme is designed based on a theme or subcategory. An extensive statistical and qualitative evaluation of the summaries generated by the system has been conducted and the results are found to be very promising.

Abs-Sum-Kan: An Abstractive Text Summarization Technique for an Indian Regional Language by Induction of Tagging Rules

This paper presents a full abstraction for Indian languages, specifically Kannada, in the context of guided summarization. The proposed process generates the abstractive sum-mary by focusing on a unified presentation model with aspect based Information Extrac-tion (IE) rules and scheme based Templates. TF/IDF rules are used for classification into categories. Lexical analysis (like Parts Of Speech tagging and Named Entity Recognition) reduces prolixity, which leads to robust IE rules. Usage of Templates for sentence genera-tion makes the summaries succinct and information intensive. The IE rules are designed to accommodate the complexities of the considered languages. Later, the system aims to produce a guided summary of domain specific documents. An abstraction scheme is a collection of aspects and associated IE rules. Each abstraction scheme is designed based on a theme or subcategory. An extensive statistical and qualitative evaluation of the summaries generated by the system has been conducted and the results are found to be very promising.

Automatic information extraction from unstructured mammography reports using distributed semantics

To date, the methods developed for automated extraction of information from radiology reports are mainly rule-based or dictionary-based, and, therefore, require substantial manual effort to build these systems. Recent efforts to develop automated systems for entity detection have been undertaken, but little work has been done to automatically extract relations and their associated named entities in narrative radiology reports that have comparable accuracy to rule-based methods. Our goal is to extract relations in a unsupervised way from radiology reports without specifying prior domain knowledge. We propose a hybrid approach for information extraction that combines dependency-based parse tree with distributed semantics for generating structured information frames about particular findings/abnormalities from the free-text mammography reports. The proposed IE system obtains a F1-score of 0.94 in terms of completeness of the content in the information frames, which outperforms a state-of-the-art rule-based system in this domain by a significant margin. The proposed system can be leveraged in a variety of applications, such as decision support and information retrieval, and may also easily scale to other radiology domains, since there is no need to tune the system with hand-crafted information extraction rules.

Information Extraction of High Resolution Remote Sensing Images Based on the Calculation of Optimal Segmentation Parameters.

Multi-scale image segmentation and the selection of optimal segmentation parameters are the key processes in the object-oriented information extraction of high-resolution remote sensing images. The accuracy of remote sensing special subject information depends on this extraction. On the basis of WorldView-2 high-resolution data, the optimal segmentation parameters methodof object-oriented image segmentation and high-resolution image information extraction, the following processes were conducted in this study. Firstly, the best combination of the bands and weights was determined for the information extraction of high-resolution remote sensing image. An improved weighted mean-variance method was proposed andused to calculatethe optimal segmentation scale. Thereafter, the best shape factor parameter and compact factor parameters were computed with the use of the control variables and the combination of the heterogeneity and homogeneity indexes. Different types of image segmentation parameters were obtained according to the surface features. The high-resolution remote sensing images were multi-scale segmented with the optimal segmentation parameters. Ahierarchical network structure was established by setting the information extraction rules to achieve object-oriented information extraction. This study presents an effective and practical method that can explain expert input judgment by reproducible quantitative measurements. Furthermore the results of this procedure may be incorporated into a classification scheme.

On learning web information extraction rules with TANGO

The research on Enterprise Systems Integration focuses on proposals to support business processes by re-using existing systems. Wrappers help re-use web applications that provide a user interface only. They emulate a human user who interacts with them and extracts the information of interest in a structured format. In this article, we present TANGO, which is our proposal to learn rules to extract information from semi-structured web documents with high precision and recall, which is a must in the context of Enterprise Systems Integration. It relies on an open catalogue of features that helps map the input documents into a knowledge base in which every DOM node is represented by means of HTML, DOM, CSS, relational, and user-defined features. Then a procedure with many variation points is used to learn extraction rules from that knowledge base; the variation points include heuristics that range from how to select a condition to how to simplify the resulting rules. We also provide a systematic method to help re-configure our proposal. Our exhaustive experimentation proves that it beats others regarding effectiveness and is efficient enough for practical purposes. Our proposal was devised to be as configurable as possible, which helps adapt it to particular web sites and evolve it when necessary.

Statistical and Analytical Study of Guided Abstractive Text Summarization

The process of creating condensed version of given text document by collecting only the important information in it is called abstractive summarization. This involves structuring the information into sentences which are simple and easy to understand. This article presents the analytical study of the process that generates abstractive summary using unified model with attribute based information extraction (IE) rules and class based templates. Classification of the document into several categories is achieved by term frequency/ inverse document frequency (TF/IDF) rules. To generate the information intensive summaries, we use templates for sentence generation. The IE rules are designed to address the complexities involved in Indian regional languages. This paper statistically analyzes the adaptation of the methodology over multiple Indian languages and many document categories. Comparisons between abstractive and extractive summaries are also presented.

A Visual and Results-Driven Rules Composition Approach for Better Information Extraction

We present a highly visual process for creating and combining elementary information extraction rules, based on their results, in order to find the rules combination that produces the most accurate information extraction results. A rule's accuracy is determined by its F-Score which is the harmonic mean of the precision and the recall of that rule. Rules are combined using logical OR and AND operators. Running a few hundreds rules combinations over a corpus, in order to determine their accuracies, can take days. Using our approach, millions of rules combinations can be tested and their accuracies (F-Score) can be calculated in few seconds. A prototype was created to demonstrate the effectiveness of our approach

A class of neural-network-based transducers for web information extraction

The Web is a huge and still growing information repository that has attracted the attention of many companies. Many such companies rely on information extractors to integrate information that is buried into semi-structured web documents into automatic business processes. Many information extractors build on extraction rules, which can be handcrafted or learned using supervised or unsupervised techniques. The literature provides a variety of techniques to learn information extraction rules that build on ad hoc machine learning techniques. In this paper, we propose a hybrid approach that explores the use of standard machine-learning techniques to extract web information. We have specifically explored using neural networks; our results show that our proposal outperforms three state-of-the-art techniques in the literature, which opens up quite a new approach to information extraction.

An Adaptive Web Information Extraction Approach Based on STU-DOM Tree

An adaptive web information extraction approach is presented in this paper. Most of the traditional web information extraction approaches depend on the templates of web sites. If the templates are changed, the information extraction rules should be redesigned. To reduce the maintenance costs and improve the adaptability of information extractors, an adaptive web information extraction approach is proposed based on the STU-DOM tree. The webpage is parsed into DOM Trees based on HTML Parser. Then DOM trees are filtered into STU-DOM trees to confirm blocks which contain keywords of a certain topic. The proposed approach is applied to webpages and the results show that the approach not only extracts information efficiently, but also is irrelevant to site structures.

Extracting Semantic Frames from Thai Medical-Symptom Unstructured Text with Unknown Target-Phrase Boundaries

Due to the limitations of language-processing tools for the Thai language, pattern-based information extraction from Thai documents requires supplementary techniques. Based on sliding-window rule application and extraction filtering, we present a framework for extracting semantic information from medical-symptom phrases with unknown boundaries in Thai unstructured-text information entries. A supervised rule learning algorithm is employed for automatic construction of information extraction rules from hand-tagged training symptom phrases. Two filtering components are introduced: one uses a classification model to predict rule application across a symptom-phrase boundary based on instantiation features of rule internal wildcards, the other uses weighted classification confidence to resolve conflicts arising from overlapping extractions. In our experimental study, we focus our attention on two basic types of symptom phrasal descriptions: one is concerned with abnormal characteristics of some observable entities and the other with human-body locations at which primitive symptoms appear. The experimental results show that the filtering components improve precision while preserving recall satisfactorily.

Automatic rule refinement for information extraction

Rule-based information extraction from text is increasingly being used to populate databases and to support structured queries on unstructured text. Specification of suitable information extraction rules requires considerable skill and standard practice is to refine rules iteratively, with substantial effort. In this paper, we show that techniques developed in the context of data provenance, to determine the lineage of a tuple in a database, can be leveraged to assist in rule refinement. Specifically, given a set of extraction rules and correct and incorrect extracted data, we have developed a technique to suggest a ranked list of rule modifications that an expert rule specifier can consider. We implemented our technique in the SystemT information extraction system developed at IBM Research -- Almaden and experimentally demonstrate its effectiveness.

A semantic approach to a framework for business domain software systems

In enterprise firms, enormous amounts of electronic documents are generated by business analysts and other business domain application users. Applications that use these documents are often driven by business logic that is hard-coded together with application logic. One approach to the separation of business logic from applications is to create and maintain business and information extraction rules in an external, user-friendly format. The drawback of such an externalization is that the business rules, usually, do not have machine interpretable semantics. This situation often leads to misinterpretation of domain analysis documents, which can inhibit the productivity of computer-assisted analytical work and the effectiveness of business solutions. This paper proposes an ontology and rule-based framework for the development of business domain applications, which includes semantic processing of externalized business rules and to some extent externalization of application logic. The creation of external information extraction rules by the business analyst is a cumbersome and time consuming task. In order to overcome this problem, the framework also includes a rule learning system to semi-automate the generation of information extraction rules from source documents with the help of manual annotations. The main idea behind the work presented in this paper is to re-engineer very large enterprise information systems to adapt to Semantic Web computing techniques. The work presented in this paper is inspired by an industrial project.

Learning to extract relations for protein annotation

Protein annotation is a task that describes protein X in terms of topic Y. Usually, this is constructed using information from the biomedical literature. Until now, most of literature-based protein annotation work has been done manually by human annotators. However, as the number of biomedical papers grows ever more rapidly, manual annotation becomes more difficult, and there is increasing need to automate the process. Recently, information extraction (IE) has been used to address this problem. Typically, IE requires pre-defined relations and hand-crafted IE rules or annotated corpora, and these requirements are difficult to satisfy in real-world scenarios such as in the biomedical domain. In this article, we describe an IE system that requires only sentences labelled according to their relevance or not to a given topic by domain experts. We applied our system to meet the annotation needs of a well-known protein family database; the results show that our IE system can annotate proteins with a set of extracted relations by learning relations and IE rules for disease, function and structure from only relevant and irrelevant sentences.

Learning rules for information extraction

The growing availability of textual sources has lead to an increase in the use of automatic knowledge acquisition approaches from textual data, as in Information Extraction (IE). Most IE systems use knowledge explicitly represented as sets of IE rules usually manually acquired. Recently, however, the acquisition of this knowledge has been faced by applying a huge variety of Machine Learning (ML) techniques. Within this framework, new problems arise in relation to the way of selecting and annotating positive examples, and sometimes negative ones, in supervised approaches, or the way of organizing unsupervised or semi-supervised approaches. This paper presents a new IE-rule learning system that deals with these training set problems and describes a set of experiments for testing this capability of the new learning approach.

An XML-enabled data extraction toolkit for web sources

The amount of useful semi-structured data on the web continues to grow at a stunning pace. Often interesting web data are not in database systems but in HTML pages, XML pages, or text files. Data in these formats are not directly usable by standard SQL-like query processing engines that support sophisticated querying and reporting beyond keyword-based retrieval. Hence, the web users or applications need a smart way of extracting data from these web sources. One of the popular approaches is to write wrappers around the sources, either manually or with software assistance, to bring the web data within the reach of more sophisticated query tools and general mediator-based information integration systems. In this paper, we describe the methodology and the software development of an XML-enabled wrapper construction system—XWRAP for semi-automatic generation of wrapper programs. By XML-enabled we mean that the metadata about information content that are implicit in the original web pages will be extracted and encoded explicitly as XML tags in the wrapped documents. In addition, the query-based content filtering process is performed against the XML documents. The XWRAP wrapper generation framework has three distinct features. First, it explicitly separates tasks of building wrappers that are specific to a web source from the tasks that are repetitive for any source, and uses a component library to provide basic building blocks for wrapper programs. Second, it provides inductive learning algorithms that derive or discover wrapper patterns by reasoning about sample pages or sample specifications. Third and most importantly, we introduce and develop a two-phase code generation framework. The first phase utilizes an interactive interface facility to encode the source-specific metadata knowledge identified by individual wrapper developers as declarative information extraction rules. The second phase combines the information extraction rules generated at the first phase with the XWRAP component library to construct an executable wrapper program for the given web source.