Guest Editorial:Special issue on trustworthy machine learning for behavioural and social computing

Machine learning has been extensively applied in behavioural and social computing,encompassing a spectrum of applications such as social network analysis,click stream analysis,recommendation of points of interest,and sentiment analysis.The datasets pertinent to these applications are inherently linked to human behaviour and societal dynamics,posing a risk of disclosing personal or sensitive information if mishandled or subjected to attacks.

Topic-Aware Abstractive Summarization Based on Heterogeneous Graph Attention Networks for Chinese Complaint Reports

Automatic text summarization(ATS)plays a significant role in Natural Language Processing(NLP).Abstractive summarization produces summaries by identifying and compressing the most important information in a document.However,there are only relatively several comprehensively evaluated abstractive summarization models that work well for specific types of reports due to their unstructured and oral language text characteristics.In particular,Chinese complaint reports,generated by urban complainers and collected by government employees,describe existing resident problems in daily life.Meanwhile,the reflected problems are required to respond speedily.Therefore,automatic summarization tasks for these reports have been developed.However,similar to traditional summarization models,the generated summaries still exist problems of informativeness and conciseness.To address these issues and generate suitably informative and less redundant summaries,a topic-based abstractive summarization method is proposed to obtain global and local features.Additionally,a heterogeneous graph of the original document is constructed using word-level and topic-level features.Experiments and analyses on public review datasets(Yelp and Amazon)and our constructed dataset(Chinese complaint reports)show that the proposed framework effectively improves the performance of the abstractive summarization model for Chinese complaint reports.

Observer-Based Adaptive Robust Precision Motion Control of a Multi-Joint Hydraulic Manipulator

Hydraulic manipulators are usually applied in heavy-load and harsh operation tasks.However,when faced with a complex operation,the traditional proportional-integral-derivative(PID)control may not meet requirements for high control performance.Model-based full-state-feedback control is an effective alternative,but the states of a hydraulic manipulator are not always available and reliable in practical applications,particularly the joint angular velocity measurement.Considering that it is not suitable to obtain the velocity signal directly from differentiating of position measurement,the low-pass filtering is commonly used,but it will definitely restrict the closed-loop bandwidth of the whole system.To avoid this problem and realize better control performance,this paper proposes a novel observerbased adaptive robust controller(obARC)for a multi-joint hydraulic manipulator subjected to both parametric uncertainties and the lack of accurate velocity measurement.Specifically,a nonlinear adaptive observer is first designed to handle the lack of velocity measurement with the consideration of parametric uncertainties.Then,the adaptive robust control is developed to compensate for the dynamic uncertainties,and the close-loop system robust stability is theoretically proved under the observation and control errors.Finally,comparative experiments are carried out to show that the designed controller can achieve a performance improvement over the traditional methods,specifically yielding better control accuracy owing to the closed-loop bandwidth breakthrough,which is limited by low-pass filtering in fullstate-feedback control.

Value of improved nursing measures and enhanced nursing management to reduce the occurrence of adverse events in pediatric infusion

BACKGROUND Intravenous infusion is a common method of drug administration in clinical practice.Errors in any aspect of the infusion process,from the verification of medical orders,preparation of the drug solution,to infusion by nursing staff,may cause adverse infusion events.AIM To analyzed the value of improving nursing measures and enhancing nursing management to reduce the occurrence of adverse events in pediatric infusion.METHODS The clinical data of 130 children who received an infusion in the pediatric department of our hospital from May 2020 to May 2021 were analyzed and divided into two groups according to the differences in nursing measures and nursing management:65 patients in the control group received conventional nursing and nursing management interventions,while 65 patients in the observation group received improved nursing measure interventions and enhanced nursing management.The occurrence of adverse events,compliance of children,satisfaction of children’s families,and complaints regarding the transfusion treatment were recorded in both groups.RESULTS The incidence of fluid extravasation and infusion set dislodgement in the observation group were 3.08%and 1.54%,respectively,which were significantly lower than 12.31%and 13.85%in the control group(P<0.05),while repeated punctures and medication addition errors in the observation group were 3.08%and 0.00%,respectively,which were lower than 9.23%and 3.08%in the control group,but there was no significant difference(P>0.05).The compliance rate of children in the observation group was 98.46%(64/65),which was significantly higher than 87.69%(57/65)in the control group,and the satisfaction rate of children’s families was 96.92%(63/65),which was significantly higher than 86.15%(56/65)in the control group(P<0.05).The observation group did not receive any complaints from the child’s family,whereas the control group received four complaints,two of which were due to the crying of the child caused by repeated punctures,one due to the poor attitude of the nurse,and one due to medication addition errors,with a cumulative complaint rate of 6.15%.The cumulative complaint rate of the observation group was significantly lower than that of the control group(P<0.05).CONCLUSION Improving nursing measures and enhancing nursing management can reduce the incidence of fluid extravasation and infusion set dislodgement in pediatric patients,improve children’s compliance and satisfaction of their families,and reduce family complaints.

Does the Nature of Chief Complaint, Gender, or Age Affect Time to Be Seen in the Emergency Room

Objectives: EM physicians may be biased in seeing patients presenting with nonspecific complaints or requiring more extensive work-ups. The goal of our study was to ascertain if chief complaint affected time to be seen (TTBS) in the ED. Methods: A retrospective report was generated from the EMR for all moderate acuity patients who visited the ED from January 2005 to December 2010 at a large urban teaching hospital. Abdominal pain, alcohol intoxication, back pain, chest pain, cough, dyspnea, dizziness, fall, fever, flank pain, headache, infection, pain (nonspecific), psychiatric evaluation, “sent by MD”, vaginal bleeding, vomiting, and weakness were the most common complaints. Non-Parametric Independent Sample Tests assessed median TTBS between complaints, gender, and age. Chisquare testing assessed for differences in the distribution of arrival times. Results: We obtained data from 116,194 patients. Patients presenting with weakness and dizziness waited the longest time of 35 minutes and patients with flank pain waited the shortest with 24 minutes. Males waited 30 minutes and females waited 32 minutes. Younger females between the ages of 18 - 50 waited significantly longer when presenting with a chief complaint of abdominal pain, chest pain, or flank pain. There was no difference in the distribution of arrival times for these complaints. Conclusion: There is a significant bias toward seeing young male patients more quickly than women or older males. Patients might benefit from efforts to educate EM physicians on the delays and potential quality issues associated with this bias in an attempt to move toward more egalitarian patient selection.

基于MacBERT和联合注意力增强网络的物业服务投诉分类方法

基于人工的物业投诉文件分类处理方法已经无法满足社会需求,并且已有投诉相关的自动分类方法在物业投诉分类问题上的性能较不足。因此,本研究提出一个基于MacBERT和联合注意力增强网络的物业服务投诉分类方法JAE BERT4Com。JAE BERT4Com使用基于近义词替换与合成少数过采样技术结合的样本增强策略解决类不平衡的问题,以及基于MacBERT的分层注意力、Transformers的多头注意力和关键词注意力等多重注意力联合增强的网络进行文本特征学习和分类。实验结果表明,JAE BERT4Com能够获得比现有模型更高的准确率、F1分数和召回率,比现有较先进模型的性能更优。

Ending Privacy’s Gremlin: Stopping the Data-Broker Loophole to the Fourth Amendment’s Search Warrant Requirement

Advances in technology require upgrades in the law. One such area involves data brokers, which have thus far gone unregulated. Data brokers use artificial intelligence to aggregate information into data profiles about individual Americans derived from consumer use of the internet and connected devices. Data profiles are then sold for profit. Government investigators use a legal loophole to purchase this data instead of obtaining a search warrant, which the Fourth Amendment would otherwise require. Consumers have lacked a reasonable means to fight or correct the information data brokers collect. Americans may not even be aware of the risks of data aggregation, which upends the test of reasonable expectations used in a search warrant analysis. Data aggregation should be controlled and regulated, which is the direction some privacy laws take. Legislatures must step forward to safeguard against shadowy data-profiling practices, whether abroad or at home. In the meantime, courts can modify their search warrant analysis by including data privacy principles.

基于KeystoneⅡ构架的Cortex A15内核系统设计与异常处理策略

阐述以高性能SOC TMS320C66AK2H的Cortex A15内核为例,利用Cortex A15内核中断控制器(GIC 400)和SOC中断处理控制器(CIC2)实现一个系统异常处理机制,能够准确快速的定位系统运行出错的地方。以Cortex A15内核使用EDMA3模块非法访问共享内存MSMC为例,验证系统异常处理机制的可行性,经测试该方法准确有效。

A mixed method analysis of patients' complaints: Underpinnings of theory-guided strategies to improve quality of care

Purpose:Patients'complaints can be predictors of patient care quality and safety.Understanding patients'complaints could help healthcare organizations target the areas for improvements.The purpose of this study is to use a mixed method analysis to a)examine the characteristics and categories of patients'complaints,b)explore the relationships of patients'complaints with professions and units,and c)propose theory-based strategies to improve care quality.Methods:This is a descriptive mixed method study.Data examined are patients'complaints filed at a university-affiliated hospital in China from January 2016 to December 2017.A qualitative content analysis was conducted to categorize complaints.A TwoStep cluster analysis was performed to provide an overall profile of patients'complaints.Chi-Square tests were conducted to investigate the relationships among complaints,professions,and units.Results:838 complaints were filed,with 821 valid cases for analysis.Six categories surfaced from the qualitative analysis:uncaring attitudes,unsatisfactory quality of treatment or competence,communication problems,the process of care,fees and billing issues,and other miscellaneous causes.Physicians received most of the complaints(56.6%).The unit receiving the most complaints were outpatient clinics and medical support units(52.7%).The cluster analysis indicated four distinct clusters.Significant relationships existed between complaints and professions(x2(20)=178.82,P<0.01),and between complaints and units(x2(15)=42.72,P<0.01).Conclusions:Patients'complaints are valuable sources for quality improvements.Healthcare providers should be not only scientifically knowledgeable,but also humanistic caring.Caring-based theories may provide guidance in clinical practice.

Relationship between Stress and Psychosomatic Complaints among Nurses in Tabarjal Hospital

This research was conducted to examine the relationship between stress and psychosomatic complaints among nurses in Tabarjal hospital. To achieve this aim, the researcher used the descriptive method and selected a sample consisting of (56) nurses working in hospital through the simple random sampling technique. The respondents were requested to complete a questionnaire for rating of stress and the most common psychosomatic diseases. The collected data were analyzed statistically by using SPSS. The results revealed that the stress was dominant among nurses in Tabarjal hospital, the prevalence of psychosomatic complaints was significantly higher in nurses, and there was significant correlation between stress and psychosomatic complaints among nurses. Conclusion: This research confirmed the effect of strains on the prevalence of psychosomatic symptoms among nurses in Tabarjal hospital. Further, implications of the results were discussed on the light of some related researches.

Quantitative Assessment of the Risk Linked to the Consumption of Braised Beef Meat “<i>Choukouya</i>” Contaminated with Pathogenic <i>Clostridium perfringens</i>in Côte D’Ivoire

Inadequate handling, processing and sales of meat products have resulted in microbial contamination responsible for most infectious human food-borne illness. The study conducted a stochastic assessment of the risk of infection linked to the consumption of braised beef meat “Choukouya” contaminated by Clostridium perfringens in C?te d’Ivoire. We conducted “Choukouya” consumers (n = 900) and vendors (n = 300) survey to characterize the actors behavior. 189 samples of “Choukouya” were collected and microbiologically analyzed according to French standard protocols NF V 08-061 and XPV 08-061. A risk model was developed and the risk of infection linked to the consumption of “Choukouya” was estimated by Monte Carlo simulation procedure. The consumer’s surveys showed that the percentage of population consuming “Choukouya” was 74.4%, with an average consumption of 114.3 ± 0.5 g/person/intake. The microbiological analysis revealed the isolation of 70% of Clostridium perfringens in the vegetative form and 60.3% in sporulated form with an average loading of 3.7 ± 2.6 log10·cfu/g and 1.1 ± 1.0 log10·cfu/g respectively. The probability of ingesting a dose greater than 109 bacteria of Clostridium prefringens varied between 7.36% and 7.93%. The braised beef meat “Choukouya” sold in the streets of cities in C?te d’Ivoire represents a real risk of infection, and Clostridium perfringens is one of the causes. This risk could be mitigated by the establishment of good hygiene practices and adequate handling processes in this informal sector.

Frequency of Hemorrhoidal Complaints in a Real-Life Population and Possible Concomitance between Hemorrhoidal Disease and Chronic Venous Disease: Going Further in Our Understanding of Hemorrhoidal Disease

Introduction: Hemorrhoids, more commonly known as piles, are swollen veins, similar to varicose veins in the lower rectum. Hemorrhoids can develop both inside and outside the rectum, and oftentimes, the cause of hemorrhoids remains unknown. This is an extremely common ailment faced by three out of four adults in their lifetime. Hemorrhoids can cause various kinds of complications, but the most common and serious ones are perianal thrombosis and incarcerated prolapsed internal hemorrhoids with subsequent thrombosis. They are accompanied by severe pain in the perianal region, and possible bleeding. Data on the coexistence of hemorrhoids with other conditions are sparse. Some data are consistent with a common pathophysiological link between straining at stool, constipation, and obstetrical events such as pregnancy and delivery. These events are also involved in the development of Chronic Vein Disease (CVD) or Chronic Vein Insufficiency (CVI). CVI is a condition that occurs when the venous walls or valves in the leg veins stop working properly, causing difficulty in blood returning to the heart from the legs. The present study was conducted with an aim to determine the frequency of complaints in hemorrhoidal patients and to assess the possible concomitance between hemorrhoidal disease and chronic venous disease. Methods: This was a multicenter, cross-sectional, observational study that enrolled patients spontaneously consulting for hemorrhoids in 17 different hospitals of different regions of Bangladesh from 1st June 2018 to 31st July 2018. Patients’ demographic and lifestyle characteristics were recorded, information on hemorrhoidal grade and signs of chronic venous disease was collected. Types of prescribed treatments were also recorded by the physicians. Results: A total of 499 patients were enrolled and analyzed. Reported frequencies of hemorrhoidal symptoms were: bleeding (80.8%), pain (66.3%), swelling (51.7%), prolapse (28.9%), itching (37.7%), soiling (12%), fecal incontinence (13.4%). 13.8% of the hemorrhoidal patients presented concomitant chronic venous disease. The Clinical-Etiological-Anatomical-Pathophysiological (CEAP) classification was used to classify the clinical signs and symptoms of the patients as C0—no visible or palpable signs of venous disease (6.2%), C1—telangiectasias or reticular veins (4.4%), C2—varicose veins (1.4%), C3—edema (2.6%), C4a—pigmentation or eczema (2%), C4b—lipodermatosclerosis or Atrophie Blanche (0.2%), C5 & C6—venous ulcer (0.6%). Commonly prescribed treatments were dietary fiber (89.8%), veno-active drugs (74.7%), topical treatments (63.7%), painkillers (11.2%), and surgical procedures (30.3%). Conclusion: The study provides small-scale data on patient profiles, risk factors, and commonly prescribed treatments in hemorrhoidal patients. The concomitance of chronic venous disease in 13.8% of the hemorrhoidal patients highlights the importance

Mediation of Refusing to Deliver Consignments——Dispute Mediation Between a Shanghai Machinery Manufacturer and a Kuwait Steel Company

Case review On March 3,2013,Mediation Center of China Council for the Promotion of International Trade(CCPIT)and China Chamber of International Commerce received a complaint lodged by a Kuwait steel company against the non-delivery of products by a Chinese machinery manufacturing company in Shanghai after collection of payment.The complaint is also copied to the Kuwait embassy in China.After receiving the complaint,the mediation center contacted with the concerned party in Kuwait and requested supporting evidence and materials.The Kuwait steel company is very satisfied with the quick feedback from the mediation center and sent

＂I＇m here to discuss my final grade!＂ A study of the speech act of complaint by Malaysian ESL learners

The present study investigates the preferred strategies of the speech act of complaint by Malaysian ESL learners. The subjects were selected from Malaysian university students majoring in English at University Sains Malaysia. Data were collected through an open-ended questionnaire in the form of a discourse completion task and a semi-structured interview. The responses were analyzed based on Rinnert and Nogami＇s （2006） taxonomy of the speech act of complaint to examine the main components of complaint, the level of directness and the degree of mitigation. It is anticipated that the findings will be able to provide valuable insights into teaching and learning English in non-native settings.

Literature and Art Move toward the World——A Visit with members of liferary and arf ctrcles at the National People’s Consultative Conference

LITERATURE and art,thespiritural products of society,are showing the effects ofreform and the socialist marketeconomy.How writers and artistsare handling the relationship be-tween art and economy is a currentissue.At the First Plenary Session ofthe Eighth National People’s Politi-cal Consultative Conference,I inter-viewed several people from China’sliterary and art circles.Facing the Reality ofChallengeSome cultural departments

An Analysis on the Complaint Speech Behavior of Online Barrage--Taking Bilibili Barrage as an Example

This study mainly combines the emerging Internet video comment mode“Barrage”to explore the main types of complaint speech acts,the object of complaint,the discourse function of complaint speech acts,and the modifiers of complaint speech acts in bullet screen culture.It is not difficult to find from previous studies on complaint speech acts that the analysis materials are mainly focused on traditional text resources and lines of film and television works,while the research objects are mainly focused on college students.Few people have studied the combination of the barrage and the complaint speech act.The results of this study show that complaints in barrage culture are mostly indirect complaint language.However,the difference between the object of complaint and the object of complaint in other traditional forms lies in that the object of complaint in direct complaint includes not only the characters in the video but also other bullet screen sender.The discourse function of complaint speech act mainly focuses on expressing reprimanding attitude and venting emotion.The modifiers of complaint speech act are mainly confirmatory modifiers.

基于Bert的电网故障处置预案信息抽取研究与实现

电网故障处置预案是及时、正确处置故障的重要依据。对其关键信息进行抽取,是实现计算机理解预案内容并进一步支撑故障处置智能化的重要基础。传统的信息抽取方法多为模式匹配,其特征和模式的获取需要专家经验,费时、费力且泛化能力较差。为实现电网故障处置预案文本信息的高效、自动抽取,文章提出一种基于Bert(bidirectional encoder representations from transformer)模型的序列标注方法。具体如下:(1)提出一种针对处置预案文本特点的数据预处理方法,该方法将句子分为不同的类别;(2)基于Bert模型自动学习和获取文本特征,并使用标注数据微调模型的方法进行信息抽取;(3)对使用上述分类方法处理过的预案文本,提出一种对每个类别分别进行模型训练并获得序列标注结果的方法。多组对比实验结果表明,该方法可以有效提升信息抽取的准确率。

基于BERT的民生问题文本分类模型——以浙江省政务热线数据为例

基于2017—2021年浙江省12345政务热线数据,从居民视角构建细粒度的民生问题三级分类体系,并利用BERT预训练模型来构建文本分类模型,将居民诉求文本转化为民生问题标签。研究结果表明,在政务热线数据中加入30%的人工生成诉求样本,可以使模型的分类准确率提升约10个百分点,准确率最高可达84.59%。对浙江省各类民生问题占比的分析结果表明,环境保护、违规经营和市政服务等诉求的比例呈现下降趋势,而公共服务、交通问题、购房问题和新兴消费模式的诉求比例呈上升趋势。研究结果有助于加强政府对于民情民意的了解,提升数据驱动的社会治理能力。

TOE Hash冲突处理设计与实现

随着以太网速率的不断提高,为了处理TCP/IP协议CPU承受了巨大的负担,因此使用单独的TCP卸载引擎处理TCP数据流显得尤为重要。TCP卸载引擎首先需要对链接进行辨认,而Hash函数作为一种发挥映射作用的函数常常被用于TCP链接的辨认。然而使用Hash函数后将不可避免地产生Hash冲突问题。设计实现的孔雀散列冲突解决方式相较传统一冗余与三冗余结构,分别节省4.1%空间与51.57%空间。在使用同种Hash且填充率为106%的情况下,孔雀散列的冲突解决率相较传统一冗余与三冗余结构分别提升21%与降低1%。在综合考虑使用空间与冲突解决率的情况下该方案有显著优势。

Best compromising crashworthiness design of automotive S-rail using TOPSIS and modified NSGAⅡ

In order to reduce both the weight of vehicles and the damage of occupants in a crash event simultaneously, it is necessary to perform a multi-objective optimal design of the automotive energy absorbing components. Modified non-dominated sorting genetic algorithm II(NSGA II) was used for multi-objective optimization of automotive S-rail considering absorbed energy(E), peak crushing force(Fmax) and mass of the structure(W) as three conflicting objective functions. In the multi-objective optimization problem(MOP), E and Fmax are defined by polynomial models extracted using the software GEvo M based on train and test data obtained from numerical simulation of quasi-static crushing of the S-rail using ABAQUS. Finally, the nearest to ideal point(NIP)method and technique for ordering preferences by similarity to ideal solution(TOPSIS) method are used to find the some trade-off optimum design points from all non-dominated optimum design points represented by the Pareto fronts. Results represent that the optimum design point obtained from TOPSIS method exhibits better trade-off in comparison with that of optimum design point obtained from NIP method.