Semi-Supervised Named Entity Recognition--论文代写范文精选

我们的工作有以下:创建一个概念验证系统，一个创新的噪声滤波技术的示范生成列表，学习消歧策略规则的验证使用自动识别,最后,缩写检测算法，解决别名解析的一种罕见且非常困难的问题。我们相信semi-supervised学习技术在机器学习社区取得了新的突破。下面的essay代写范文进行详述。

Abstract 
Named Entity Recognition (NER) aims to extract and to classify rigid designators in text such as proper names, biological species, and temporal expressions. There has been growing interest in this field of research since the early 1990s. In this thesis, we document a trend moving away from handcrafted rules, and towards machine learning approaches. Still, recent machine learning approaches have a problem with annotated data availability, which is a serious shortcoming in building and maintaining large-scale NER systems. In this thesis, we present an NER system built with very little supervision. Human supervision is indeed limited to listing a few examples of each named entity (NE) type. 

First, we introduce a proof-of-concept semi-supervised system that can recognize four NE types. Then, we expand its capacities by improving key technologies, and we apply the system to an entire hierarchy comprised of 100 NE types. Our work makes the following contributions: the creation of a proof-of-concept semisupervised NER system; the demonstration of an innovative noise filtering technique for generating NE lists; the validation of a strategy for learning disambiguation rules using automatically identified, unambiguous NEs; and finally, the development of an acronym detection algorithm, thus solving a rare but very difficult problem in alias resolution. We believe semi-supervised learning techniques are about to break new ground in the machine learning community. 

In this thesis, we show that limited supervision can build complete NER systems. On standard evaluation corpora, we report performances that compare to baseline supervised systems in the task of annotating NEs in texts.

Introduction 
The term “Named Entity” (NE) is in current use in Information Extraction (IE) applications. It was coined at the sixth Message Understanding Conference (MUC-6) (Grishman & Sundheim 1996), which influenced IE research in the 1990s. At the time, MUC was focusing on IE tasks wherein structured information on company and defense-related activities are extracted from unstructured text, such as newspaper articles. In defining IE tasks, people noticed that it is essential to recognize information units such as names including person, organization, and location names, and numeric expressions including time, date, money, and percentages. Identifying references to these entities in text was acknowledged as one of IE’s important sub-tasks and was called “Named Entity Recognition (NER).” Before the NER field was recognized in 1996, significant research was conducted by extracting proper names from texts. 

A paper published in 1991 by Lisa F. Rau (1991) is often cited as the root of the field. For more than fifteen years, a dynamic research community advanced the fundamental knowledge and the engineered solutions to create an NER system. In its canonical form, the input of an NER system is a text and the output is information on boundaries and types of NEs found in the text. The vast majority of proposed systems fall in two categories: the handmade rule-based systems; and the supervised learning-based systems. In both approaches, large collections of documents are analyzed by hand to obtain sufficient knowledge for designing rules or for feeding machine learning algorithms. Expert linguists must execute this important amount of work, which in turn limits the building and maintenance of large-scale NER systems. This thesis is about the creation of an autonomous NER system. It has the desirable property of requiring a small amount of work by an expert linguist. It falls in the new category of semi-supervised and unsupervised systems. Influential work in this category is relatively rare and recent, and we believe ours to be the first thesis devoted exclusively to the creation of an autonomous NER system.

Discussion and Conclusion 
This thesis is about creating a semi-supervised NER system. It has the desirable property of requiring, as input, that an expert linguist lists a dozen examples of each supported entity type. It contrasts with the annotation of thousands of documents with hundreds of entity types, which is required for supervised learning. It also contrasts with manually harvesting NE lists and designing a complex rule system, which are usually required for handmade systems. The NER system we present in this thesis therefore requires very little supervision and we’ve included this human input in the Appendix. The system presented in this thesis falls in the new category of semi-supervised and unsupervised systems. Work in this category is relatively rare and recent, and we believe ours to be the first that is devoted exclusively to the autonomous creation of an NER system. 

Our overall goal is to create proof-of-concept software. In completing this system, we claim four major contributions that impact the NER field, and also have the potential to be used in other domains. First, we designed the first semi-supervised NER system that performs at a comparable level to that of a simple supervised learning-based NER system (Chapter 3). Second, we present a noise filter for generating NE lists based on computational linguistics and statistical semantic techniques (Chapter 4). This noise filter outperforms previous systems devoted to the same task. Then, we demonstrate a simple technique based on set intersections that can identify unambiguous examples for a given NE type (Chapter 5). 

Unambiguous NEs are a requirement for creating semi-supervised disambiguation rules. Finally, our fourth contribution is an acronym detection algorithm—part of an alias resolution system—that outperforms previous system and allows improvement in NER for a “less common and very difficult problem” (Chapter 6). These contributions are crucial components to a successful semi-supervised NER system, and they are explained in the context of the whole system, for which the architecture is detailed in Figure 1. In the course of completing this system, however, we met many 111 limitations and difficulties, which we discuss in Section 7.1. We conclude this thesis by presenting our future work and some general long-term research ideas. We believe the resulting system requiring little supervision has two important advantages over past systems put forth in the literature, and this is generally in favour of a shift towards semi-supervised and unsupervised techniques in the machine learning community. Our system is first extensible to new entity types. 

The design we adopted is free of linguistic knowledge or type-dependant heuristics. Therefore, we can modify the hierarchy or add new types, and let the system generate lists and rules. The system is also easily maintained over time. While supervised learning-based systems get most of their knowledge from large static training corpora, the system we present gets most of its knowledge from the Web. Recrawling the Web and periodically verifying the Web pages from which lists were extracted is a straightforward approach to maintenance.(论文代写)

51Due网站原创范文除特殊说明外一切图文著作权归51Due所有；未经51Due官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，51Due保留一切法律追诉权。(论文代写)
更多论文代写范文欢迎访问我们主页 www.51due.com 当然有论文代写需求可以和我们24小时在线客服 QQ:800020041 联系交流。-X(论文代写)