基于多视角的多类型错误全面检测方法

随着信息化社会的发展,数据的规模越发庞大,数据的种类也越发丰富.时至今日,数据已经成为国家和企业的重要战略资源,是科学化管理的重要保障.然而,随着社会生活产生的数据日益丰富,大量的脏数据也随之而来,数据质量问题油然而生.如何准确而全面地检测出数据集中所包含的错误数据,一直是数据科学中的痛点问题.尽管已有许多传统方法被广泛用于各行各业,如基于约束与统计的检测方法,但这些方法通常需要丰富的先验知识与昂贵的人力和时间成本.受限于此,这些方法往往难以准确而全面地检测数据.近年来,许多新型错误检测方法利用深度学习技术,通过时序推断、文本解析等方式取得了更好检测效果,但它们通常只适用于特定的领域或特定的错误类型,面对现实生活中的复杂情况,泛用性不足.基于上述情况,结合传统方法与深度学习技术的优点,提出了一个基于多视角的多类型错误全面检测模型CEDM.首先,从模式的角度,结合现有约束条件,在属性、单元和元组层面进行多维度的统计分析,构建出基础检测规则;然后,通过词嵌入捕获数据语义,从语义的角度分析属性相关性、单元关联性与元组相似性,进而基于语义关系,从多个维度上更新、扩展基础规则;最终,联合多个视角...     

基于实时视频感知的虚拟体育交互系统

针对疫情常态化背景下,传统体育项目受场地、器材等限制,市场上相关产品价格昂贵、可扩展性不足等问题,提出了一种基于实时视频感知的虚拟体育交互系统.该系统设计视频数据采集模块和人体关节点提取模块,结合OpenPose获取人体的关节点坐标,实时捕捉人体手势以及肢体动作.动作语义理解模块包括运动动作理解和绘图动作理解.前者根据运动中肢体关节点的相对位置关系,识别运动动作语义.后者将手腕部关节点绘图动作轨迹生成为草图图像,使用AlexNet进行识别分类,解析为对应的绘制动作语义.该模型在边缘端设备的分类准确率为98.83%.采用基于Unity设计的草图游戏应用作为可视化交互界面,实现在虚拟场景中的运动交互.该系统使用实时视频感知交互方式实现居家运动健身,无需其他的外部设备,具有更强的参与度和趣味性.     

基于K Framework的向量化机器学习指令语义形式化

ARM针对ARMv8.1-M微处理器架构推出基于M-Profile向量化扩展方案的技术,并命名为ARM Helium,声明能为ARM Cortex-M处理器提升达15倍的机器学习性能.随着物联网的高速发展,微处理器指令执行正确性尤为重要.指令集的官方手册作为芯片模拟程序,片上应用程序开发的依据,是程序正确性基本保障.主要介绍利用可执行语义框架K Framework对ARMv8.1-M官方参考手册中向量化机器学习指令的语义正确性研究.基于ARMv8.1-M的官方参考手册自动提取指令集中描述向量化机器学习指令执行过程的伪代码,并将其转换为形式化语义转换规则.通过K Framework提供的可执行框架利用测试用例,验证机器学习指令算数运算执行的正确性.     

Measuring code maintainability with deep neural networks

The maintainability of source code is a key quality characteristic for software quality. Many approaches have been proposed to quantitatively measure code maintainability. Such approaches rely heavily on code metrics, e.g., the number of Lines of Code and McCabe’s Cyclomatic Complexity. The employed code metrics are essentially statistics regarding code elements, e.g., the numbers of tokens, lines, references, and branch statements. However, natural language in source code, especially identifiers, is rarely exploited by such approaches. As a result, replacing meaningful identifiers with nonsense tokens would not significantly influence their outputs, although the replacement should have significantly reduced code maintainability. To this end, in this paper, we propose a novel approach (called DeepM) to measure code maintainability by exploiting the lexical semantics of text in source code. DeepM leverages deep learning techniques (e.g., LSTM and attention mechanism) to exploit these lexical semantics in measuring code maintainability. Another key rationale of DeepM is that measuring code maintainability is complex and often far beyond the capabilities of statistics or simple heuristics. Consequently, DeepM leverages deep learning techniques to automatically select useful features from complex and lengthy inputs and to construct a complex mapping (rather than simple heuristics) from the input to the output (code maintainability index). DeepM is evaluated on a manually-assessed dataset. The evaluation results suggest that DeepM is accurate, and it generates the same rankings of code maintainability as those of experienced programmers on 87.5% of manually ranked pairs of Java classes.     

基于语义的单文档自动摘要算法

单文档自动摘要的目的是在原始的文本中通过摘取、提炼主要信息,提供一篇简洁全面的摘要。自动摘要的主流方法是通过统计和机器学习的技术从文本中直接提取出句子,而单文档由于篇章有限,统计的方法无效。针对此问题,提出了基于语义的单文本自动摘要方法。该方法首先将文档划分为句子,然后计算每一对句子的语义相似度,通过运用改进型K-Medoids聚类算法将相似的句子归类,在每一类中选出最具代表性的句子,最后将句子组成文档摘要。实验结果表明,通过融合语义信息,该方法提高了摘要的质量。     

Defining the meaning of tabular mathematical expressions

Mathematical expressions in tabular form (also called “tabular expressions” or “tables”) have been shown to be useful for documenting and analysing software systems. They are usually easier to read than conventional mathematical expressions but are no less precise. They can be used wherever mathematical expressions are used. To avoid misunderstandings, and to support users with trustworthy tools, the meaning of these expressions must be fully defined.This paper presents a new method for defining the meaning of tabular expressions. Each definition of an expression type names the expression’s constituents, and provides a restriction schema and one or more evaluation schemas. The restriction schema defines the class of well-formed expressions of the type. An evaluation schema maps a well-formed tabular expression of the type to a mathematical expression of a previously defined type. Since the meaning of conventional mathematical expressions is well known, describing an equivalent expression fully defines the meaning of a tabular expression.In this approach, indexation is used to decouple the appearance of a tabular expression from its semantics. A tabular expression is an indexed set of grids; a grid is an indexed set of expressions. The expressions in a grid can be either conventional expressions or tabular expressions of a previously defined type.Defining the meaning of a tabular expression in this way facilitates the building of tools that faithfully implement the semantics. The decoupling of syntax and semantics by means of indices overcomes some limitations of older approaches.The method presented in the paper is illustrated by defining several previously known types of tabular expressions and some new ones. The use of the new model to build a suite of tools for the input, presentation, validation, evaluation, simplification, conversion and composition of tabular expressions is discussed.     

灯具形态语义设计的编码与解码研究

文章将符号学中的编码解码理论应用于灯具形态设计中,从形状、色彩、材质三方面分析了灯具形态的语义特点,重点研究了灯具形态语义编码与解码的方式与特点。     

Bialgebras for structural operational semantics: An introduction

Bialgebras and distributive laws are an abstract, categorical framework to study various flavors of structural operational semantics. This paper aims to introduce the reader to the basics of bialgebras for operational semantics, and to sketch the state of the art in this research area.