Comparing automated text classification methods

Online social media drive the growth of unstructured text data. Many marketing applications require structuring this data at scales non-accessible to human coding, e.g., to detect communication shifts in sentiment or other researcher-defined content categories. Several methods have been proposed to automatically classify unstructured text. This paper compares the performance of ten such approaches (five lexicon-based, five machine learning algorithms) across 41 social media datasets covering major social media platforms, various sample sizes, and languages. So far, marketing research relies predominantly on support vector machines (SVM) and Linguistic Inquiry and Word Count (LIWC). Across all tasks we study, either random forest (RF) or naive Bayes (NB) performs best in terms of correctly uncovering human intuition. In particular, RF exhibits consistently high performance for three-class sentiment, NB for small samples sizes. SVM never outperform the remaining methods. All lexicon-based approaches, LIWC in particular, perform poorly compared with machine learning. In some applications, accuracies only slightly exceed chance. Since additional considerations of text classification choice are also in favor of NB and RF, our results suggest that marketing research can benefit from considering these alternatives.     

中国旅游景区功能演变与用地分类构想

研究目的:基于中国旅游景区功能演变、用地特征及问题分析,构建旅游景区用地分类体系,以期为旅游景区用地纳入区域土地利用提供理论基础,为旅游景区规划的深度编制提供实践依据。研究方法:通过调研和问卷厘清现状景区用地情况,对比借鉴相关用地分类体系,基于此构建旅游景区用地分类方案。研究结果:分析并阐明了旅游景区的功能演变、用地特征和现状问题,构建了2大类、9中类、28小类的景区用地分类体系,并与《土地利用现状分类》进行衔接。研究结论:建立可衔接且具可操作性的旅游景区用地分类体系,是实现旅游景区健康可持续发展与用地规范化管控的关键。     

Machine learning for financial transaction classification across companies using character-level word embeddings of text fields

An important initial step in accounting is mapping financial transfers to the corresponding accounts. We devised machine-learning-based systems that automate this process. They use word embeddings with character-level features to process transaction texts. When considering 473 companies independently, our approach achieved an average top-1 accuracy of 80.50%, outperforming baselines that exclude the transaction texts or rely on a lexical bag-of-words text representation. We extended the approach to generalizes across companies and even across different corporate sectors. After standardization of the account structures and careful feature engineering, a single classifier trained on 44 companies from 28 sectors achieved a test accuracy of more than 80%. When trained on 43 companies and tested on the remaining one, the system achieved an average performance of 64.62%. This rate increased to nearly 70% when considering only the largest sector.     

Research on a diagnostic system of rural vitalization based on development elements in China

In 2017, the Chinese government implemented a national strategy of "Rural Vitalization" that sought to realize full-scale rural vitalization. However, is it possible to achieve vitalization for all the villages in China? How should their development potential be determined? This paper identified and analyzed the "element-composite" messages of rural development based on 99 exemplary sites of “Beautiful Villages” in China. Combined with the projection pursuit classification method, a diagnostic system of rural vitalization was established; then, Dehua County was taken as a case study for an in-depth analysis. Based on national data analysis, the final results indicated that livelihood resources (LR), agglomeration effects (AE), location and transportation (LT), cultural/natural landscapes (CN), and economic circumstance (EC) are essential elements for successful rural development. Additionally, EC was the only exogenous element, while the remaining elements were endogenous. Furthermore, the villages with better EC presented urbanization rates of 38∼82 % and Engel coefficients of 29∼41 % in their counties; exemplary sites lacking LR, CN, LT, and AE account for 13.13 %, 19.19 %, 26.26 %, and 60.61 % respectively, so the indispensability of these elements decreases progressively in sequence. Only 2 % of villages rely on single element for success, therefore, the composite pattern of development element was also critical; 10 out of 16 types were found to successfully facilitate village development, among which, the type of R-a-L-C (32.32 %) and R-A-L-C (15.15 %) were considered as the greatest potential patterns for vitalization. Finally, by means of the diagnostic system, the ratio of representative villages for high-low potential in Dehua County is evenly split; then, development paths, and land use policies that match with paths were proposed, on the basis of development potential and “element-composite” condition of themselves.     

The role of text-extracted investor sentiment in Chinese stock price prediction with the enhancement of deep learning

Whether investor sentiment affects stock prices is an issue of long-standing interest for economists. We conduct a comprehensive study of the predictability of investor sentiment, which is measured directly by extracting expectations from online user-generated content (UGC) on the stock message board of Eastmoney.com in the Chinese stock market. We consider the influential factors in prediction, including the selections of different text classification algorithms, price forecasting models, time horizons, and information update schemes. Using comparisons of the long short-term memory (LSTM) model, logistic regression, support vector machine, and Naïve Bayes model, the results show that daily investor sentiment contains predictive information only for open prices, while the hourly sentiment has two hours of leading predictability for closing prices. Investors do update their expectations during trading hours. Moreover, our results reveal that advanced models, such as LSTM, can provide more predictive power with investor sentiment only if the inputs of a model contain predictive information.     

Waste of money or growth opportunity: The causal effect of EU subsidies on Hungarian SMEs

Although EU subsidies aiming at economic development play a pivotal role not only for Hungary but for the entire European Union as well, there is a debate regarding their effectiveness in the literature. This paper investigates the impact of direct economic development subsidies extended in the context of Structural Funds and the Cohesion Fund as part of the 2007–2013 programming period of the European Union on Hungarian micro, small and medium-sized enterprises. Based on a micro database, we evaluate the impact of corporates’ first subsidies on various performance indicators, using a combination of propensity score matching and fixed effects panel regression. According to our results, economic development funds had a significant positive effect on the number of employees, sales revenue, gross value added and, in some cases, operating profit. However, the labour productivity of enterprises was not significantly affected by any of the support schemes. Furthermore, by explicitly comparing non-refundable subsidies (grants) and refundable assistance (financial instruments), we find that there is no significant difference in the effectiveness of the two types of subsidy.     

基于深度学习的跳频信号识别

针对跳频信号分选存在人工提取参数特征具有复杂性的问题，提出了一种基于深度学习的识别方法。首先对跳频信号进行短时傅里叶变换，得到二维的时频矩阵；接着提取信号的轮廓特征，构造三维矩阵作等高线图，并对等高线图进行预处理；最后把预处理后的等高线图输入到卷积神经网络中进行训练、测试，进而实现分类识别。仿真结果表明，在不需要复杂的人工提取参数特征的基础上，在分选率为100〖WT《Times New Roman》〗%〖WTBZ〗时，所提方法经裁剪处理下的信噪比为-15 dB，比支持向量机和传统K-Means聚类算法都低10 dB。实测数据的算法验证表明，所提方法能够将大疆精灵4Pro、hm无人机、司马航模X8HW以及大疆悟2这四类无人机正确分类。     

一种基于微波雷达回波信号的车型分类方法

车辆类型识别方法是智能交通系统的关键技术之一。利用深度学习的高维特征泛化学习能力，将改进的LeNet-5卷积神经网络用于基于交通微波雷达的大小车型分类识别。首先，以雷达触发前的N帧信号为基础，对雷达的回波信号进行分析并构建数据集；然后，分析LeNet-5卷积神经网络的特点；最后提出一种改进的LeNet-5卷积神经网络。实验结果表明，与传统的支持向量机方法相比，所提方法能够智能学习大小车的雷达时频信号特征，大小车型识别准确率达到97%以上，可为交通场景下的车型识别研究提供新的技术途径。     

关于改进个人银行账户分类管理问题的分析

随着互联网和移动通讯技术的发展,支付领域风险范围、扩散速度、溢出效应程几何级数增长。为有效防范网络新型违法犯罪,保护人民群众财产安全和合法权益,中国人民银行相继发布了多项制度条例,全面推进个人账户分类管理。本文结合对商业银行账户分类管理专项调查结果,分析分类管理中存在的问题及原因,在此基础上,提出进一步改进个人银行账户分类管理的措施和建议。     

深度学习方法在作物遥感分类中的应用和挑战

[目的]准确估算作物的面积和分布对粮食安全至关重要。与传统的机器学习方法相比,深度学习具有多种优势,如端到端训练、可迁移性。为有效利用高时空数据进行作物识别提供了新的机遇。已有多种模型被应用于作物分类任务中,针对不同的分类任务,如何有效地选择模型,并对其进行训练和使用已成为关键问题。[方法]文章回顾了利用深度学习模型对作物分类的主要研究。N维卷积神经网络(N-D CNN)(N=1、2、3)和递归神经网络(RNN)已被有效用于作物分类任务。长短期记忆RNN(LSTM RNN)和门控循环单元RNN(GRU RNN)是RNN的变体,解决了随着时间序列增加RNN出现的梯度消失或爆炸问题。此外,还有研究使用CNN和RNN(我们称为RCNN)的混合模型对作物进行分类。该文首先阐述了使用深度学习方法进行作物制图的背景和意义,并介绍了CNN和RNN模型结构。然后回顾了一些典型的研究,包括模型的结构、遥感数据源、数据处理方法和分类精度。最后,总结了使用深度学习方法进行作物分类的挑战以及现有解决方案的局限性。[结果](1)1-D CNN可用于提取时间特征,或时间+光谱特征,分类效果良好;2-D CNN已被广泛应用于单时相数据的空间特征提取,分类精度依赖于数据源;3-D CNN应用较少,但具有很大的潜力,尤其是时间+空间维度的特征提取;(2)相同条件下(架构、数据源、研究区域、类别),LSTM RNN和GRU RNN分类效果通常高于普通RNN,而前两者的效果差距不大,但GRU RNN训练时间较短;(3)CNN+RNN混合模型(RCNN)用RNN比3-D CNN更适合提取时间特征。这主要是由于RNN建立了对序列数据的长期依赖,而3-D CNN卷积核是局部计算的。[结论]通过分析,认为深度学习技术是作物遥感分类的有效工具。此外,与其他模型相比,RCNN,3-D CNN和GUR RNN具有更大的潜力。