基于深度学习的中文分词方法研究

作为中文自然语言处理中的基础任务中文分词,其分词的好坏直接影响之后的自然语言处理任务。当前中文分词大部分都是采用基于机器学习的方法,但是其需要人工构建大量特征。针对上述问题,论文提出一种基于深度学习的新分词模型,该模型基于BLSTM(双向长短期神经网络),CNN(卷积神经网络)和CRF(条件随机场),充分利用了BLSTM可以利用长距离信息和CNN提取局部信息的优点。并设计了实验,在三个数据集上验证论文提出的模型在中文分词上的正确性和优越性。     

University Learning and Anti-Plagiarism Back-End Services

Plagiarism refers to the use of other people’s ideas and information without acknowledging the source. In this research, anti-plagiarism software was designed especially for the university and its campuses to identify plagiarized text in students’ written assignments and laboratory reports. The proposed framework collected original documents to identify plagiarized text using natural language processing. Our research proposes a method to detect plagiarism by applying the core concept of text, which is semantic associations of words and their syntactic composition. Information on the browser was obtained through Request application programming interface by name Url.AbsoluteUri, and it is stored in a centralized Microsoft database Server. A total of 55,001 data samples were collected from 2015 to 2019. Furthermore, we assimilated data from a university website, specifically from the psau.edu.sa network, and arranged the data into students’ categories. Furthermore, we extracted words from source documents and student documents using the WordNet library. On a benchmark dataset consisting of 785 plagiarized text and 4,716 original text data, a significant accuracy of 90.2% was achieved. Therefore, the proposed framework demonstrated better performance than the other available tools. Many students mentioned that working on assignments using the framework was suitable because they were able to work on the assignments in harmony, as per their timeframe and from different network locations. The framework also recommends procedures that can be used to avoid plagiarism.     

A superstructure‐based model for multistream heat exchanger design within flow sheet optimization

Multistream heat exchangers (MHEXs) are often used in energy‐intensive cryogenic processes. Modeling them within a process optimization formulation has been a challenge due to the needs to accommodate phase changes and ensure temperature approach. In this work, we present a nonlinear model for MHEXs based on a novel single‐stage superstructure of two‐stream exchangers. Our formulation guarantees a minimum temperature approach for all heat exchanges, estimates heat exchange areas for individual stream matches, requires no prior knowledge of phase changes, uses no Boolean variables, and enables seamless optimization of a process with multiple MHEXs. Furthermore, it facilitates dedicated constant‐phase intervals that allow accurate estimation of heat‐transfer parameters for various stream matches. We optimize two natural gas liquefaction processes involving MHEXs, and report better solutions than the existing literature. © 2017 American Institute of Chemical Engineers AIChE J, 63: 3764–3777, 2017     

Normalization of informal text

This paper describes a noisy-channel approach for the normalization of informal text, such as that found in emails, chat rooms, and SMS messages. In particular, we introduce two character-level methods for the abbreviation modeling aspect of the noisy channel model: a statistical classifier using language-based features to decide whether a character is likely to be removed from a word, and a character-level machine translation model. A two-phase approach is used; in the first stage the possible candidates are generated using the selected abbreviation model and in the second stage we choose the best candidate by decoding using a language model. Overall we find that this approach works well and is on par with current research in the field.     

关系抽取方法研究

关系抽取是自然语言处理技术的重要基础之一,也是机器学习、模式识别和本体工程等相关领域广泛研究的问题。近年来,随着相关领域中新技术的不断涌现,关系抽取方法也得到了新发展。对这些新发展进行了较详细的归纳,总结了关系抽取方法发展的趋势。     

BP 网络日径流序列预测的NLP 适用条件研究

对非线性预处理在人工神经网络日径流预测中的适应过程进行了仿真和模拟．提出了非线性预处理（NLP）适用条件的解算思路，通过实测数据和模拟数据，研究了NLP的适用条件。推导出NLP在神经网络SISO系统中适合于日径流预测，不适用于周平均流量序列、旬平均流量序列和月平均流量序列的预测，提出了判断NLP神经网络SISO系统进行日径流预测的有效性标准——多年日径流拐点14百分位．并通过广西平乐水文站和四川宝珠寺水文站1973～2001年的日径流量进行对比预测，验证了该标准是合理的。     

关于自然语言计算机处理的几点思考

自然语言处理是人类在信息时代探索自我的必然,为此应将其看作为人类认识自我进程中的一个环节.本文结合其他学科,简要探讨了有关语言和智能,语言是否可以计算,语义和分解,歧义与知识等基本问题.     

A Hybrid Intelligent Approach for Content Authentication and Tampering Detection of Arabic Text Transmitted via Internet

In this paper, a hybrid intelligent text zero-watermarking approach has been proposed by integrating text zero-watermarking and hidden Markov model as natural language processing techniques for the content authentication and tampering detection of Arabic text contents. The proposed approach known as Second order of Alphanumeric Mechanism of Markov model and Zero-Watermarking Approach (SAMMZWA). Second level order of alphanumeric mechanism based on hidden Markov model is integrated with text zero-watermarking techniques to improve the overall performance and tampering detection accuracy of the proposed approach. The SAMMZWA approach embeds and detects the watermark logically without altering the original text document. The extracted features are used as a watermark information and integrated with digital zero-watermarking techniques. To detect eventual tampering, SAMMZWA has been implemented and validated with attacked Arabic text. Experiments were performed on four datasets of varying lengths under multiple random locations of insertion, reorder and deletion attacks. The experimental results show that our method is more sensitive for all kinds of tampering attacks with high level accuracy of tampering detection than compared methods.     

利用语音的上升过零间隔的语音识别研究

本文对语音的上升过零间隔从发音机理及统计角度进行了分析,认为语音信号及其差分信号的上升过零间隔较好地体现了不同语音之间的差异性,是一种进行语音识别的理想参数,并利用该参数建立模板,以非线性分块法进行时间对准,以相邻三帧最佳匹配法进行模板匹配,用软件模拟了特定人、小词汇量、孤立词的语音识别,得到了较好的实验结果。     

基于超文本的概念词典管理系统

本文介绍了自然语言处理系统的一个重要知识源——概念词典,以及基于超文本的概念词典管理系统——概念浏览器。我们的概念词典以框架结构为知识表示单元,框架的内容由结构信息和语义信息组成,所有框架通过结构槽形成树形的层次结构。概念浏览器作为一个超文本管理系统,不仅能显示层次结构和框架内容,还提供了插入、查询、删除和修改等各种数据库功能。