The impact of deep learning on document classification using semantically rich representations

This paper presents a semantically rich document representation model for automatically classifying financial documents into predefined categories utilizing deep learning. The model architecture consists of two main modules including document representation and document classification. In the first module, a document is enriched with semantics using background knowledge provided by an ontology and through the acquisition of its relevant terminology. Acquisition of terminology integrated to the ontology extends the capabilities of semantically rich document representations with an in depth-coverage of concepts, thereby capturing the whole conceptualization involved in documents. Semantically rich representations obtained from the first module will serve as input to the document classification module which aims at finding the most appropriate category for that document through deep learning. Three different deep learning networks each belonging to a different category of machine learning techniques for ontological document classification using a real-life ontology are used.Multiple simulations are carried out with various deep neural networks configurations, and our findings reveal that a three hidden layer feedforward network with 1024 neurons obtain the highest document classification performance on the INFUSE dataset. The performance in terms of F1 score is further increased by almost five percentage points to 78.10% for the same network configuration when the relevant terminology integrated to the ontology is applied to enrich document representation. Furthermore, we conducted a comparative performance evaluation using various state-of-the-art document representation approaches and classification techniques including shallow and conventional machine learning classifiers.     

A scalable and adaptive method for finding semantically equivalent cue words of uncertainty

Scientific knowledge is constantly subject to a variety of changes due to new discoveries, alternative interpretations, and fresh perspectives. Understanding uncertainties associated with various stages of scientific inquiries is an integral part of scientists’ domain expertise and it serves as the core of their meta-knowledge of science. Despite the growing interest in areas such as computational linguistics, systematically characterizing and tracking the epistemic status of scientific claims and their evolution in scientific disciplines remains a challenge. We present a unifying framework for the study of uncertainties explicitly and implicitly conveyed in scientific publications. The framework aims to accommodate a wide range of uncertainty types, from speculations to inconsistencies and controversies. We introduce a scalable and adaptive method to recognize semantically equivalent cues of uncertainty across different fields of research and accommodate individual analysts’ unique perspectives. We demonstrate how the new method can be used to expand a small seed list of uncertainty cue words and how the validity of the expanded candidate cue words is verified. We visualize the mixture of the original and expanded uncertainty cue words to reveal the diversity of expressions of uncertainty. These cue words offer a novel resource for the study of uncertainty in scientific assertions.     

Developing a topic-driven method for interdisciplinarity analysis

This study explores the topic-based interdisciplinarity in the research domain of literacy. A text corpus of keywords was generated through a deep keyword generation model from abstracts of 346,387 articles published in 296 disciplines from 1917 to 2021. Dirichlet-Multinomial Regression topic modeling, interdisciplinarity indices, and network analysis were employed to analyze the collected corpus. Topic modeling uncovered 15 dominant research topics in the literacy field, as well as their up-and-down trends from 2000 to 2021. For each topic, keywords were then replaced with disciplines, and interdisciplinarity was measured using four indices: variety, balance, disparity, and diversity. Finally, the interdisciplinarity of each topic, connectivity between topics, and topic trends were comprehensively analyzed on the keyword co-occurrence network. Our methodology reaches beyond connectivity limited to a few disciplines and provides insight into the direction of collaboration between disciplines centered on a research domain. Moreover, the study's deep keyword generation model has methodological implications for forming a corpus spanning numerous disciplines as a bottom-up approach.     

“戏子”之痛——张恨水笔下的女性抒写微探

作为张恨水社会言情小说着重描写的对象之一,＂戏子＂的形象几乎贯穿于其小说创作的始终。作为社会一个特殊的群体,在艺术还只是供人消遣而尚未受到人们充分尊重的年代里,艺人的悲剧命运值得人们去反思。通过对《啼笑因缘》和《夜深沉》的个案分析,来探讨张恨水笔下的戏子形象及其造成其悲剧命运的自身因素,并由此来考察张恨水小说在暴露国民性弱点和鞭策民族劣根性、重塑民族魂方面所作出的努力与尝试。     

Should music evoke emotion?-Review on Deep Concern in New Horizon College English

Deep Concern in New Horizon College English (Reading and Writing Book1) describes a generation gap in an American family. In the text, the parents worry about their daughter’s listening to rock music for it may bring negative impact on her. Nowadays there is definitely music with unwholesome or depraved thoughts that may mislead listeners. Should people therefore refuse to listen to music or refuse the influence of music on emotion? The essay will focus on the issue to have a review on the relationship between emotion and music.     

校企深度融合的探索

"以校企合作体制机制创新为核心,全面深化校企合作"是未来中国高职教育发展的必由之路。由于多种复杂的因素,目前高职院校在校企合作中表现出诸多的矛盾和不适应。就校企深度融合的制约因素、途径和措施,结合本校实际情况进行分析和研究,旨在探索校企深度融合的发展之路。     

朱自清散文的艺术特色及成因探究

在中国现代文坛上朱自清作为＂散文圣手＂而被后来者广为赞誉。他的散文感情真、语言美、结构巧、道理深,被誉为现代＂白话美文的模范＂。探究其成因,源于＂厚爱＂,正是朱自清对家人、对人民、对祖国、对大自然、对艺术创作的厚爱铸就了他散文创作的艺术丰碑。     

基于通用搜索引擎的深层网络表面化方法研究

在现有相关研究的基础上,对基于通用搜索引擎的深层网络表面化方法的基本原理进行分析,对表单域取值范围的确定、查询处理、查询结果的超链接设置等与深层网络表面化相关的若干关键问题进行探讨。     

采用最小DFS的Deep Web结构化数据抽取

通过分析动态数据在其Web 页面中的展示特点,提出一个新的自动化、结构化数据抽取方法。首先基于DOM利用算法实现快速定位数据区,从而避免处理大量噪音数据;其次引入最小DFS编码来表示DOM子树,通过聚类对记录数据区进行区分;最后对少量样本页面训练学习生成抽取规则用于数据抽取。利用原型系统针对实际网站中的页面进行数据抽取,实验结果显示其拥有较高的准确性和效率。     

强制性运动疗法及“白脉散”有效成分组对实验性MCAO／R模型神经血管单元细胞凋亡的影响及机制研究

目的：研究强制性运动疗法、“白脉散”抗脑缺血有效成分纽及二者联用对大鼠脑缺血再灌注损伤的保护作用及相关机制。方法：75只雄性SD大鼠随机分为假手术组（Sham）、模型组（MCAO／R）、银杏叶提取物组（EGB）、强制性运动疗法组（CIMT）、“白脉散”有效成分组（ECG）,强制性运动疗法＋“白脉散”有效成分组（CIMT＋ECG）,每组7只。采用线栓法制作大鼠大脑中动脉闭塞再灌注模型,缺血2h再灌注24h后进行神经功能评分,然后ECG组及cIMT＋ECG组以155mg／kg／d灌胃,EGB组以5．6mg／kg／d灌胃,其余组灌胃等体积的溶媒,同时CIMT组及CIMT＋ECG组建立运动疗法模型,每天1次,连续7天。利用免疫组化法观察神经血管单元（NVU）细胞中Bcl-2、Caspase-3蛋白表达的情况,运用原位末端标记（TUNEL）法计数NVU凋亡细胞。结果：CIMT与“白脉散”ECG单独或联合应用可显著改善模型大鼠神经功能评分、抑制NVU细胞凋亡、增加Bcl-2蛋白的表达,降低Caspase-3蛋白的表达。结论：CIMT及“白脉散”ECG可保护MCAO／R模型大鼠神经功能,其机制可能与其抗NVU细胞凋亡,调节凋亡调控因子蛋白表达有关,且二者联用比二者单用具有更佳的保护效果。