Automatic classification of construction safety reports using semi-supervised YAKE-Guided LDA approach

Literature on supervised Machine-Learning (ML) approaches for classifying text-based safety reports for the construction sector has been growing. Recent studies have emphasized the need to build ML approaches that balance high classification accuracy and performance on management criteria, such as resource intensiveness. However, despite being highly accurate, the extensively focused, supervised ML approaches may not perform well on management criteria as many factors contribute to their resource intensiveness. Alternatively, the potential for semi-supervised ML approaches to achieve balanced performance has rarely been explored in the construction safety literature. The current study contributes to the scarce knowledge on semi-supervised ML approaches by demonstrating the applicability of a state-of-the-art semi-supervised learning approach, i.e., Yet, Another Keyword Extractor (YAKE) integrated with Guided Latent Dirichlet Allocation (GLDA) for construction safety report classification. Construction-safety-specific knowledge is extracted as keywords through YAKE, relying on accessible literature with minimal manual intervention. Keywords from YAKE are then seeded in the GLDA model for the automatic classification of safety reports without requiring a large quantity of prelabeled datasets. The YAKE-GLDA classification performance (F1 score of 0.66) is superior to existing unsupervised methods for the benchmark data containing injury narratives from Occupational Health and Safety Administration (OSHA). The YAKE-GLDA approach is also applied to near-miss safety reports from a construction site. The study demonstrates a high degree of generality of the YAKE-GLDA approach through a moderately high F1 score of 0.86 for a few categories in the near-miss data. The current research demonstrates that, unlike the existing supervised approaches, the semi-supervised YAKE-GLDA approach can achieve a novel possibility of consistently achieving reasonably good classification performance across various construction-specific safety datasets yet being resource-efficient. Results from an objective comparative and sensitivity analysis contribute to much-required knowledge-contesting insights into the functioning and applicability of the YAKE-GLDA. The results from the current study will help construction organizations implement and optimize an efficient ML-based knowledge-mining strategy for domains beyond safety and across sites where the availability of a pre-labeled dataset is a significant limitation.     

On-site text classification and knowledge mining for large-scale projects construction by integrated intelligent approach

A large-scale project produces a lot of text data during construction commonly achieved as various management reports. Having the right information at the right time can help the project team understand the project status and manage the construction process more efficiently. However, text information is presented in unstructured or semi-structured formats. Extracting useful information from such a large text warehouse is a challenge. A manual process is costly and often times cannot deliver the right information to the right person at the right time. This research proposes an integrated intelligent approach based on natural language processing technology (NLP), which mainly involves three stages. First, a text classification model based on Convolution Neural Network (CNN) is developed to classify the construction on-site reports by analyzing and extracting report text features. At the second stage, the classified construction report texts are analyzed with improved frequency-inverse document frequency (TF-IDF) by mutual information to identify and mine construction knowledge. At the third stage, a relation network based on the co-occurrence matrix of the knowledge is presented for visualization and better understanding of the construction on-site information. Actual construction reports are used to verify the feasibility of this approach. The study provides a new approach for handling construction on-site text data which can lead to enhancing management efficiency and practical knowledge discovery for project management.     

基于WordNet概念向量空间模型的文本分类

文章提出了一种文本特征提取方法,以WordNet语言本体库为基础,以同义词集合概念代替词条,同时考虑同义词集合间的上下位关系,建立文本的概念向量空间模型作为文本特征向量,使得在训练过程中能够提取出代表类别的高层次信息。实验结果表明,当训练文本集合很小时,方法能够较大地提高文本的分类准确率。     

An effective and interpretable method for document classification

As the number of documents has been rapidly increasing in recent time, automatic text categorization is becoming a more important and fundamental task in information retrieval and text mining. Accuracy and interpretability are two important aspects of a text classifier. While the accuracy of a classifier measures the ability to correctly classify unseen data, interpretability is the ability of the classifier to be understood by humans and provide reasons why each data instance is assigned to a label. This paper proposes an interpretable classification method by exploiting the Dirichlet process mixture model of von Mises–Fisher distributions for directional data. By using the labeled information of the training data explicitly and determining automatically the number of topics for each class, the learned topics are coherent, relevant and discriminative. They help interpret as well as distinguish classes. Our experimental results showed the advantages of our approach in terms of separability, interpretability and effectiveness in classification task of datasets with high dimension and complex distribution. Our method is highly competitive with state-of-the-art approaches.     

HeadlineStanceChecker: Exploiting summarization to detect headline disinformation

The headline of a news article is designed to succinctly summarize its content, providing the reader with a clear understanding of the news item. Unfortunately, in the post-truth era, headlines are more focused on attracting the reader’s attention for ideological or commercial reasons, thus leading to mis- or disinformation through false or distorted headlines. One way of combating this, although a challenging task, is by determining the relation between the headline and the body text to establish the stance. Hence, to contribute to the detection of mis- and disinformation, this paper proposes an approach (HeadlineStanceChecker) that determines the stance of a headline with respect to the body text to which it is associated. The novelty rests on the use of a two-stage classification architecture that uses summarization techniques to shape the input for both classifiers instead of directly passing the full news body text, thereby reducing the amount of information to be processed while keeping important information. Specifically, summarization is done through Positional Language Models leveraging on semantic resources to identify salient information in the body text that is then compared to its corresponding headline. The results obtained show that our approach achieves 94.31% accuracy for the overall classification and the best FNC-1 relative score compared with the state of the art. It is especially remarkable that the system, which uses only the relevant information provided by the automatic summaries instead of the whole text, is able to classify the different stance categories with very competitive results, especially in the discuss stance between the headline and the news body text. It can be concluded that using automatic extractive summaries as input of our approach together with the two-stage architecture is an appropriate solution to the problem.     

基于深度学习的文本分类综述

随着互联网的不断发展,网络上的文本数据日益增多,如果能对这些数据进行有效分类,那么更有利于从中挖掘出有价值的信息,因此文本数据的管理和整合显得十分重要。文本分类是自然语言处理任务中的一项基础性工作,主要应用于舆情检测及新闻文本分类等领域,目的是对文本资源进行整理和归类。基于深度学习的文本分类,在对文本数据处理中,表现出较好的分类效果。本文对用于文本分类的深度学习算法进行详细阐述,按照深度学习的不同算法进行分类,并分析各种算法的特点,最后对深度学习算法在文本分类领域的未来研究方向进行总结。     

Exploring the fine-grained analysis and automatic detection of irony on Twitter

To push the state of the art in text mining applications, research in natural language processing has increasingly been investigating automatic irony detection, but manually annotated irony corpora are scarce. We present the construction of a manually annotated irony corpus based on a fine-grained annotation scheme that allows for identification of different types of irony. We conduct a series of binary classification experiments for automatic irony recognition using a support vector machine (SVM) that exploits a varied feature set and compare this method to a deep learning approach that is based on an LSTM network and (pre-trained) word embeddings. Evaluation on a held-out corpus shows that the SVM model outperforms the neural network approach and benefits from combining lexical, semantic and syntactic information sources. A qualitative analysis of the classification output reveals that the classifier performance may be further enhanced by integrating implicit sentiment information and context- and user-based features.     

基于领域特征文本的Deep Web分类研究

Deep Web自动分类是建立深网数据集成系统的前提和基础。提出了一种基于领域特征文本的Deep Web分类方法。首先借助本体知识对表达同一语义的不同词汇进行了概念抽象,进而给出了领域相关度的定义,并将其作为特征文本选择的量化标准,避免了人为选取的主观性和不确定性;在接口向量模型构建中,考虑了不同特征文本对于分类作用的差异,提出了一种改进的W-TFIDF权重计算方法;最后采用KNN算法对接口向量进行了分类。对比实验证明,利用所提方法选择的特征文本是准确有效的,新的特征文本权重计算方法能显著地提高分类精度,且在KNN算法中表现出较好的稳定性。     

An innovative hybrid approach for extracting named entities from unstructured text data

Named entity recognition (NER) is the core part of information extraction that facilitates the automatic detection and classification of entities in natural language text into predefined categories, such as the names of persons, organizations, locations, and so on. The output of the NER task is crucial for many applications, including relation extraction, textual entailment, machine translation, information retrieval, etc. Literature shows that machine learning and deep learning approaches are the most widely used techniques for NER. However, for entity extraction, the abovementioned approaches demand the availability of a domain‐specific annotated data set. Our goal is to develop a hybrid NER system composed of rule‐based deep learning as well as clustering‐based approaches, which facilitates the extraction of generic entities (such as person, location, and organization) out of natural language texts of domains that lack generic named entities labeled domain data sets. The proposed approach takes the advantages of both deep learning and clustering approaches but separately, in combination with a knowledge‐based approach by using a postprocessing module. We evaluated the proposed methodology on court cases (judgments) as a use case since it contains generic named entities of different forms that are poorly or not present in open‐source NER data sets. We also evaluated our hybrid models on two benchmark data sets, namely, Computational Natural Language Learning (CoNLL) 2003 and Open Knowledge Extraction (OKE) 2016. The experimental results obtained from benchmark data sets show that our hybrid models achieved substantially better performance in terms of the F‐score in comparison to other competitive systems.     

Text-based approaches for non-topical image categorization

The rapid expansion of multimedia digital collections brings to the fore the need for classifying not only text documents but their embedded non-textual parts as well. We propose a model for basing classification of multimedia on broad, non-topical features, and show how information on targeted nearby pieces of text can be used to effectively classify photographs on a first such feature, distinguishing between indoor and outdoor images. We examine several variations to a TF*IDF-based approach for this task, empirically analyze their effects, and evaluate our system on a large collection of images from current news newsgroups. In addition, we investigate alternative classification and evaluation methods, and the effects that secondary features have on indoor/outdoor classification. Using density estimation over the raw TF*IDF values, we obtain a classification accuracy of 82%, a number that outperforms baseline estimates and earlier, image-based approaches, at least in the domain of news articles, and that nears the accuracy of humans who perform the same task with access to comparable information. Published online: 22 September 2000     

Automated classification of software change messages by semi-supervised Latent Dirichlet Allocation

ContextTopic models such as probabilistic Latent Semantic Analysis (pLSA) and Latent Dirichlet Allocation (LDA) have demonstrated success in mining software repository tasks. Understanding software change messages described by the unstructured nature-language text is one of the fundamental challenges in mining these messages in repositories.ObjectiveWe seek to present a novel automatic change message classification method characterized by semi-supervised topic semantic analysis.MethodIn this work, we present a semi-supervised LDA based approach to automatically classify change messages. We use domain knowledge of software changes to make labeled samples which are added to build the semi-supervised LDA model. Next, we verify the cross-project analysis application of our method on three open-source projects. Our method has two advantages over existing software change classification methods: First of all, it mitigates the issue of how to set the appropriate number of latent topics. We do not have to choose the number of latent topics in our method, because it corresponds to the number of class labels. Second, this approach utilizes the information provided by the label samples in the training set.ResultsOur method automatically classified about 85% of the change messages in our experiment and our validation survey showed that 70.56% of the time our automatic classification results were in agreement with developer opinions.ConclusionOur approach automatically classifies most of the change messages which record the cause of the software change and the method is applicable to cross-project analysis of software change messages.     

Using deep learning and social network analysis to understand and manage extreme flooding

Combining machine learning with social network analysis (SNA) can leverage vast amounts of social media data to better respond to crises. We present a case study using Twitter data from the March 2019 Nebraska floods in the United States, which caused over $1 billion in damage in the state and widespread evacuations of residents. We use a subset of machine learning, deep learning (DL), to classify text content of 11,982 tweets, and we integrate that with SNA to understand the structure of tweet interactions. Our DL approach pre‐trains our model with a DL language technique, BERT, and then trains the model using the standard training dataset to sort a dataset of tweets into classes tailored to crisis events. Several performance measures demonstrate that our two‐tiered trained model improves domain adaptation and generalization across different extreme weather event types. This approach identifies the role of Twitter during the damage containment stage of the flood. Our SNA identifies accounts that function as primary sources of information on Twitter. Together, these two approaches help crisis managers filter large volumes of data and overcome challenges faced by simple statistical models and other computational techniques to provide useful information during crises like flooding.     

Hazard analysis: A deep learning and text mining framework for accident prevention

Learning from past accidents is pivotal for improving safety in construction. However, hazard records are typically documented and stored as unstructured or semi-structured free-text rendering the ability to analyse such data a difficult task. The research presented in this study presents a novel and robust framework that combines deep learning and text mining technologies that provide the ability to analyse hazard records automatically. The framework comprises four-step modelling approach: (1) identification of hazard topics using a Latent Dirichlet Allocation algorithm (LDA) model; (2) automatic classification of hazards using a Convolution Neural Network (CNN) algorithm; (3) the production of a Word Co-occurrence Network (WCN) to determine the interrelations between hazards; and (4) quantitative analysis by Word Cloud (WC) technology of keywords to provide a visual overview of hazard records. The proposed framework is validated by analysing hazard records collected from a large-scale transport infrastructure project. It is envisaged that the use of the framework can provide managers with new insights and knowledge to better ensure positive safety outcomes in projects. The contributions of this research are threefold: (1) it is demonstrated that the process of analysing hazard records can be automated by combining deep learning and text learning; (2) hazards are able to be visualized using a systematic and data-driven process; and (3) the automatic generation of hazard topics and their classification over specific time periods enabling managers to understand their patterns of manifestation and therefore put in place strategies to prevent them from reoccurring.     

An information fusion approach to integrate image annotation and text mining methods for geographic knowledge discovery

Due to the steady increase in the number of heterogeneous types of location information on the internet, it is hard to organize a complete overview of the geospatial information for the tasks of knowledge acquisition related to specific geographic locations. The text- and photo-types of geographical dataset contain numerous location data, such as location-based tourism information, therefore defining high dimensional spaces of attributes that are highly correlated. In this work, we utilized text- and photo-types of location information with a novel approach of information fusion that exploits effective image annotation and location based text-mining approaches to enhance identification of geographic location and spatial cognition. In this paper, we describe our feature extraction methods to annotating images, and utilizing text mining approach to analyze images and texts simultaneously, in order to carry out geospatial text mining and image classification tasks. Subsequently, photo-images and textual documents are projected to a unified feature space, in order to generate a co-constructed semantic space for information fusion. Also, we employed text mining approaches to classify documents into various categories based upon their geospatial features, with the aims to discovering relationships between documents and geographical zones. The experimental results show that the proposed method can effectively enhance the tasks of location based knowledge discovery.     

一种基于混合词向量的胶囊网络文本分类方法

文本分类是目前深度学习方法被广泛应用的重要领域之一.本文设计了一种基于循环神经网络和胶囊网络的混合模型,使用胶囊网络来克服卷积神经网络对空间不敏感的缺点,学习文本局部与整体之间的关系特征,并使用循环神经网络中的GRU神经网络经过最大池化层来学习上下文显著信息特征,结合两者来优化特征提取过程从而提高文本分类效果.与此同时,在嵌入层中提出了一种基于缺失词补全的混合词向量方法,采用两种策略来减少缺失词过度匹配的现象以及降低词向量中的噪声数据出现的概率,从而获得语义丰富且少噪声的高质量词向量.在经典文本分类数据集中进行实验,通过与对比模型的最优方法进行比较,证明了该模型和方法能有效地提升文本分类准确度.     

结合Bert与Bi-LSTM的英文文本分类模型

作为自然语言处理技术中的底层任务之一,文本分类任务对于上游任务有非常重要的辅助价值。而随着最近几年深度学习广泛应用于NLP中的上下游任务的趋势,深度学习在下游任务文本分类中性能不错。但是目前的基于深层学习网络的模型在捕捉文本序列的长距离型上下文语义信息进行建模方面仍有不足,同时也没有引入语言信息来辅助分类器进行分类。针对这些问题,提出了一种新颖的结合Bert与Bi-LSTM的英文文本分类模。该模型不仅能够通过Bert预训练语言模型引入语言信息提升分类的准确性,还能基于Bi-LSTM网络去捕捉双向的上下文语义依赖信息对文本进行显示建模。具体而言,该模型主要有输入层、Bert预训练语言模型层、Bi-LSTM层以及分类器层搭建而成。实验结果表明,与现有的分类模型相比较,所提出的Bert-Bi-LSTM模型在MR数据集、SST-2数据集以及CoLA数据集测试中达到了最高的分类准确率,分别为86.2%、91.5%与83.2%,大大提升了英文文本分类模型的性能。     

Deep infrared pedestrian classification based on automatic image matting

Infrared pedestrian classification plays an important role in advanced driver assistance systems. However, it encounters great difficulties when the pedestrian images are superimposed on a cluttered background. Many researchers design very deep neural networks to classify pedestrian from cluttered background. However, a very deep neural network associated with a high computational cost. The suppression of cluttered background can boost the performance of deep neural networks without increasing their depth, while it has received little attention in the past. This study presents an automatic image matting approach for infrared pedestrians that suppresses the cluttered background and provides consistent input to deep learning. The domain expertise in pedestrian classification is applied to automatically and softly extract foreground objects from images with cluttered backgrounds. This study generates trimaps, which must be generated manually in conventional approaches, according to the estimated positions of pedestrian’s head and upper body without the need for any user interaction. We implement image matting by adopting the global matting approach and taking the generated trimap as an input. The representation of pedestrian is discovered by a deep learning approach from the resulting alpha mattes in which cluttered background is suppressed, and foreground is enhanced. The experimental results show that the proposed approach improves the infrared pedestrian classification performance of the state-of-the-art deep learning approaches at a negligible computational cost.     

基于词典和弱标注信息的电影评论情感分析

针对评论文本情感分析研究中数据标注费时费力的问题,提出了一种新的数据自动标注方法。首先,通过基于情感词典的方法计算出评论文本的情感倾向;其次,利用用户评分的弱标注信息和基于词典方法的情感倾向对评论文本自动标注;最后,利用支持向量机（SVM）对评论文本进行情感分类。所提出的数据自动标注方法在两种类型数据集情感分类准确率上分别达到了77.2%和77.8%,相对于单一的利用用户评分对数据标注的方法,分别提高了1.7个百分点和2.1个百分点。实验结果表明,提出的数据自动标注方法在电影评论情感分析中能提高分类效果。