采用多尺度注意力机制的远程监督关系抽取

针对目前大多数关系抽取模型中局部特征及全局特征利用不充分的缺点,该文提出一种采用多尺度注意力机制的远程监督关系抽取模型。在词语层面,通过在池化层构建权重矩阵来衡量词语与关系的相关程度,从而捕捉句子中重要的语义特征;在句子层面,采用注意力机制将预测关系与句子进行相关性比较,获得句子级别的重要信息。模型在NYT数据集上平均准确率达到78%,表明该模型能够有效地利用多尺度特征,并且提高远程关系抽取任务的准确率。     

An extensive analysis of the interaction between missing data types,imputation methods,and supervised classifiers

When applying data-mining techniques to real-world data, we often find ourselves facing observations that have no value recorded for some attributes. This can be caused by several phenomena, such as a machine’s incapability to record certain characteristics or a person refusing to answer a question in a poll. Depending on that motivation, values gone missing may follow one kind of pattern or another, or describe no regularity at all. One approach to palliate the effect of missing data on machine learning tasks is to replace the missing observations. Imputation algorithms attempt to calculate a value for a missing gap, using information associated with it, i.e., the attribute and/or other values in the same observation. While several imputation methods have been proposed in the literature, few works have addressed the question of the relationship between the type of missing data, the choice of the imputation method, and the effectiveness of classification algorithms that used the imputed data. In this paper we address the relationship among these three factors. By constructing a benchmark of hundreds of databases containing different types of missing data, and applying several imputation methods and classification algorithms, we empirically show that an interaction between imputation methods and supervised classification can be deduced. Besides, differences in terms of classification performance for the same imputation method in different missing data patterns have been found. This points to the convenience of considering the combined choice of the imputation method and the classifier algorithm according to the missing data type.     

水质监测无线传感器网络的硬件设计

针对传统水质监测系统不能对水质参数进行实时在线监测,难以准确检测水质参数的动态变化、水质参数检测误差大等问题,提出了水质监测无线传感器网络（ WSNs）的硬件设计方案。系统主要通过核心单片机CC2530实现传感器节点设计,采用太阳能电池板进行供电,同时设计了采集温度、pH 值的硬件电路,并对硬件电路进行了稳定性试验。在 IRA 开发环境下,进行传感器节点和协调器的编程,使之能够进行通信。实验结果表明：系统温度、pH值的平均相对误差分别为3.06%,1.64%,提高了监测精度。     

复杂细微差异化网络数据特征的语义优化提取算法

对网络数据的复杂、细微、差异化特征进行语义提取,是实现Web网络数据准确识别和检索的关键技术。复杂、细微、差异化的网络数据语义特征具有非线性和随机散布性的特点,其主题分布广、更新频率大,从而造成语义特征提取困难。传统方法采用小波基函数投影算法进行语义特征的提取,性能不好。提出了一种基于Dopplerlet变换匹配投影的网络数据特征语义优化提取算法。首先构建语义高斯边缘化矩形窗函数进行融合滤波处理,通过文本切分把大量的信息熵数据进行小波基函数投影,有效剔除簇内异常数据;然后利用Dopplerlet变换匹配投影的自相似特性,自适应匹配语义的非线性谱特征,在Hilbert张成子空间中,实现对语义特征的提取和优化表达,再完成提取。仿真实验表明,该算法提高了对网络数据特征语义的表达能力,能有效区分差异网络数据中的冗余数据和残差数据,提高对杂细微差异化网络数据的检测识别和检索能力。     

基于物联网云平台的壁挂炉远程控制系统

详细描述了基于Ayla云平台的远程壁挂炉控制系统设计方案与具体实现, 通过Android手机客户端, 接入Ayla物联网云平台, 实现对家中壁挂炉的远程控制. 壁挂炉与Ayla云平台连接的中间设备, 采用以WM-N-BM-09A片上WiFi芯片和STM32F100BV主控芯片为核心的Ayla嵌入式模块进行开发, 利用WiFi通过无线路由与Ayla云平台建立通信通道, 壁挂炉的运行状态信息、报警数据, 通过Ayla云平台的主动推送机制, 实时推送至手机客户端. 测试结果表明, 基于Ayla物联网云平台的远程控制系统方案, 成本低, 易于实现. 此方案可以被应用于其它远程智能控制系统, 为智能设备的远程控制提供了一种低成本、高可靠的解决方案.     

加入自注意力机制的BERT命名实体识别模型

命名实体识别属于自然语言处理领域词法分析中的一部分,是计算机正确理解自然语言的基础。为了加强模型对命名实体的识别效果,本文使用预训练模型BERT(bidirectional encoder representation from transformers)作为模型的嵌入层,并针对BERT微调训练对计算机性能要求较高的问题,采用了固定参数嵌入的方式对BERT进行应用,搭建了BERT-BiLSTM-CRF模型。并在该模型的基础上进行了两种改进实验。方法一,继续增加自注意力(self-attention)层,实验结果显示,自注意力层的加入对模型的识别效果提升不明显。方法二,减小BERT模型嵌入层数。实验结果显示,适度减少BERT嵌入层数能够提升模型的命名实体识别准确性,同时又节约了模型的整体训练时间。采用9层嵌入时,在MSRA中文数据集上F₁值提升至94.79%,在Weibo中文数据集上F₁值达到了68.82%。     

BLSTM-PA在食品安全领域语义分析仿真研究

食品安全领域的智能问答系统旨在对用户通过自然语言进行的食品安全方面的提问做出快速、简洁的反馈,其技术挑战主要在于语义分析和答案句子表示,尤其是在于如何消除问答之间的词汇差距以加强问答匹配能力,以及如何抓取准确的核心单词以增强句子表示能力。尽管基于"短语级别"和众多的注意力模型已经取得了一定的性能提升,但基于注意力的框架都没有很好的重视位置信息。针对上述问题,运用词林和word2vec相结合的方法,提出近义词-主词替换机制(将普通词映射为核心词),实现了语义表示的归一化。同时,受位置上下文提升信息检索性能的启发,假设如果问句中的一个词(称之为问题词)出现在答案句中,问题词的临近词对比偏离词应该被给与更高的权重。基于上述假设,提出了基于双向lstm模型的位置注意力机制(BLSTM-PA)。上述机制给与答案句中问题词的临近文本更高的注意力。以食品安全问答系统为语义分析验证和仿真的平台,通过在食品安全领域数据集(即FS-QA)上进行的对比实验,从MAP和MRR评价指标来看,与基于传统的注意力机制的RNN模型相比,BLSTM-PA实现了5.96%的提升,证明了BLSTM-PA模型的良好性能,同时,集成了提出的问答模型的食品安全问答系统性能也得到了显著的提升。     

Advertisement clicking prediction by using multiple criteria mathematical programming

In online advertisement industry, it is important to predict potentially profitable users who will click target ads (i.e., Behavioral targeting). The task selects the potential users that are likely to click the ads by analyzing user’s clicking/web browsing information and displaying the most relevant ads to them. This paper proposes four multiple criteria mathematical programming models for advertisement clicking problems. First two are multi-criteria linear regression (MCLR) and kernel-based multiple criteria regression (KMCR) algorithms for click-through rate (CTR) prediction. The second two are multi-criteria linear programming (MCLP) and kernel-based multiple criteria programming (KMCP) algorithms, which are used to predict ads clicking events, such as identifying clicked ads in a set of ads. Using the experimental datasets from KDD Cup 2012, the paper first conducts a comparison of the proposed MCLR and KMCR with the methods of support vector regression (SVR) and logistic regression (LR), which shows that both MCLR and KMCR are good alternatives. Then the paper further studies the performance between the proposed MCLP and KMCP algorithms with known algorithms, including support vector machines (SVM), LR, radial basis function network (RBFN), k-nearest neighbor algorithm (KNN) and Naïve Bayes (NB) in both prediction and selection processes. The studies show that the MCLP and KMCP models have better performance stability and can be used to effectively handle behavioral targeting application for online advertisement problems.     

A perceptual computing-based method to prioritize failure modes in failure mode and effect analysis and its application to edible bird nest farming

A failure mode and effect analysis (FMEA) procedure that incorporates a novel Perceptual Computing (Per-C)–based Risk Priority Number (RPN) model is proposed in this paper. The proposed model considers linguistic uncertainties and vagueness of words, because it is more natural to use words, instead of numerals, for an FMEA user to express his/her knowledge when he/she provides an assessment. Therefore, it is important to consider the inherited uncertainties in words used by humans for assessment as an additional risk factor in the entire FMEA reasoning process. As such, we propose to use Per-C to analyze the uncertainties in words provided by different FMEA users. There are three potential sources of risks. Firstly, the risk factors of Severity (S), Occurrence (O), and Detection (D) are graded using words by each FMEA user, and indicated as interval type-2 fuzzy sets (IT2FSs). Secondly, the relative importance of S, O, and D are reflected by the weights given by each FMEA user in words, which are indicated as IT2FSs. Thirdly, the expertise level of each FMEA user is reflected by words, which are expressed as IT2FSs too. The proposed Per-C-RPN model allows these three sources of risks from each FMEA user to be considered and combined in terms of IT2FSs. A case study related to edible bird nest farming in Borneo Island is reported. The results indicate the effectiveness of the proposed model. In summary, this paper contributes to a new Per-C-RPN model that utilizes imprecise assessment grades pertaining to group decision making in FMEA.