Driving rule extraction based on cognitive behavior analysis

In order to make full use of the driver’s long-term driving experience in the process of perception, interaction and vehicle control of road traffic information, a driving behavior rule extraction algorithm based on artificial neural network interface(ANNI) and its integration is proposed. Firstly, based on the cognitive learning theory, the cognitive driving behavior model is established, and then the cognitive driving behavior is described and analyzed. Next, based on ANNI, the model and the rule extraction algorithm(ANNI-REA) are designed to explain not only the driving behavior but also the non-sequence. Rules have high fidelity and safety during driving without discretizing continuous input variables. The experimental results on the UCI standard data set and on the self-built driving behavior data set, show that the method is about 0.4% more accurate and about 10% less complex than the common C4.5-REA, Neuro-Rule and REFNE. Further, simulation experiments verify the correctness of the extracted driving rules and the effectiveness of the extraction based on cognitive driving behavior rules. In general, the several driving rules extracted fully reflect the execution mechanism of sequential activity of driving comprehensive cognition, which is of great significance for the traffic of mixed traffic flow under the network of vehicles and future research on unmanned driving.

Rough Set Theory Based Approach for Fault Diagnosis Rule Extraction of Distribution System

As the first step of service restoration of distribution system,rapid fault diagnosis is a significant task for reducing power outage time,decreasing outage loss,and subsequently improving service reliability and safety.This paper analyzes a fault diagnosis approach by using rough set theory in which how to reduce decision table of data set is a main calculation intensive task.Aiming at this reduction problem,a heuristic reduction algorithm based on attribution length and frequency is proposed.At the same time,the corresponding value reduction method is proposed in order to fulfill the reduction and diagnosis rules extraction.Meanwhile,a Euclid matching method is introduced to solve confliction problems among the extracted rules when some information is lacking.Principal of the whole algorithm is clear and diagnostic rules distilled from the reduction are concise.Moreover,it needs less calculation towards specific discernibility matrix,and thus avoids the corresponding NP hard problem.The whole process is realized by MATLAB programming.A simulation example shows that the method has a fast calculation speed,and the extracted rules can reflect the characteristic of fault with a concise form.The rule database,formed by different reduction of decision table,can diagnose single fault and multi-faults efficiently,and give satisfied results even when the existed information is incomplete.The proposed method has good error-tolerate capability and the potential for on-line fault diagnosis.

Non-invasive prediction of non-alcoholic steatohepatitis in Japanese patients with morbid obesity by artificial intelligence using rule extraction technology

AIM To construct a non-invasive prediction algorithm for predicting non-alcoholic steatohepatitis(NASH), we investigated Japanese morbidly obese patients using artificial intelligence with rule extraction technology.METHODS Consecutive patients who required bariatric surgery underwent a liver biopsy during the operation. Standard clinical, anthropometric, biochemical measurements were used as parameters to predict NASH and were analyzed using rule extraction technology. One hundred and two patients, including 79 NASH and 23 non-NASH patients were analyzed in order to create the predictionmodel, another cohort with 77 patients including 65 NASH and 12 non-NASH patients were analyzed to validate the algorithm.RESULTS Alanine aminotransferase, C-reactive protein, homeostasis model assessment insulin resistance, albumin were extracted as predictors of NASH using a recursive-rule extraction algorithm. When we adopted the extracted rules for the validation cohort using a highly accurate rule extraction algorithm, the predictive accuracy was 79.2%. The positive predictive value, negative predictive value,sensitivity and specificity were 88.9%, 35.7%, 86.2% and 41.7%, respectively.CONCLUSION We successfully generated a useful model for predicting NASH in Japanese morbidly obese patients based on their biochemical profile using a rule extraction algorithm.

Rule Extraction： Using Neural Networks or for Neural Networks？

In the research of rule extraction from neural networks, fidelity describeshow well the rules mimic the behavior of a neural network while accuracy describes how well therules can be generalized. This paper identifies the fidelity-acuracy dilemma. It argues todistinguish rule extraction using neural networks and rule extraction for neural networks accordingto their different goals, where fidelity and accuracy should be excluded from the rule qualityevaluation framework, respectively.

Extraction Fuzzy Linguistic Rules from Neural Networks for Maximizing Tool Life in High-speed Milling Process

In metal cutting industry it is a common practice to search for optimal combination of cutting parameters in order to maximize the tool life for a fixed minimum value of material removal rate（MRR）. After the advent of high-speed milling（HSM） pro cess, lots of experimental and theoretical researches have been done for this purpose which mainly emphasized on the optimization of the cutting parameters. It is highly beneficial to convert raw data into a comprehensive knowledge-based expert system using fuzzy logic as the reasoning mechanism. In this paper an attempt has been presented for the extraction of the rules from fuzzy neural network（FNN） so as to have the most effective knowledge-base for given set of data. Experiments were conducted to determine the best values of cutting speeds that can maximize tool life for different combinations of input parameters. A fuzzy neural network was constructed based on the fuzzification of input parameters and the cutting speed. After training process, raw rule sets were extracted and a rule pruning approach was proposed to obtain concise linguistic rules. The estimation process with fuzzy inference showed that the optimized combination of fuzzy rules provided the estimation error of only 6.34 m/min as compared to 314 m/min of that of randomized combination of rule s.

Forecasting tourism demand by extracting fuzzy Takagi-Sugeno rules from trained SVMs

Tourism demand forecasting has attracted substantial interest because of the significant economic contributions of the fast-growing tourism industry. Although various quantitative forecasting techniques have been widely studied, highly accurate and understandable forecasting models have not been developed. The present paper proposes a novel tourism demand forecasting method that extracts fuzzy Takagi-Sugeno （T-S） rules from trained SVMs. Unlike previous approaches, this study uses fuzzy T-S models extracted from the outputs of trained SVMs on tourism data. Owing to the symbolic fuzzy rules and the generalization ability of SVMs, the extracted fuzzy T-S rules exhibit high forecasting accuracy and include understandable pre-condition parts for practitioners. Based on the tourism demand forecasting problem in Hong Kong SAR, China as a case study, empirical findings on tourist arrivals from nine overseas origins reveal that the proposed approach performs comparably with SVMs and can achieve better prediction accuracy than other forecasting techniques for most origins. The findings demonstrated that decision makers can easily interpret fuzzy T-S rules extracted from SVMs. Thus, the approach is highly beneficial to tourism market management. This finding demonstrates the excellent scientific and practical values of the proposed approach in tourism demand forecasting.

Rule Extraction from Trained Artificial Neural Network Based on Genetic Algorithm

This paper discusses how to extract symbolic rules from trained artificial neural network (ANN) in domains involving classification using genetic algorithms (GA). Previous methods based on an exhaustive analysis of network connections and output values have already been demonstrated to be intractable in that the scale-up factor increases with the number of nodes and connections in the network. Some experiments explaining effectiveness of the presented method are given as well.

Towards explicit representation of an artificial neural network model: Comparison of two artificial neural network rule extraction approaches

In the quest for interpretable models,two versions of a neural network rule extraction algorithm were proposed and compared.The two algorithms are called the Piece-Wise Linear Artificial Neural Network(PWL-ANN)and enhanced Piece-Wise Linear Artificial Neural Network(enhanced PWL-ANN)algorithms.The PWL-ANN algorithm is a decomposition artificial neural network(ANN)rule extraction algorithm,and the enhanced PWL-ANN algorithm improves upon the PWL-ANN algorithm and extracts multiple linear regression equations from a trained ANN model by approximating the hidden sigmoid activation functions using N-piece linear equations.In doing so,the algorithm provides interpretable models from the originally trained opaque ANN models.A detailed application case study illustrates how the generated enhanced-PWL-ANN models can provide understandable IF-THEN rules about a problem domain.Comparison of the results generated by the two versions of the PWL-ANN algorithm showed that in comparison to the PWL-ANN models,the enhanced-PWL-ANN models support improved fidelities to the originally trained ANN models.The results also showed that more concise rule sets could be generated using the enhanced-PWL-ANN algorithm.If a more simplified set of rules is desired,the enhanced-PWL-ANN algorithm can be combined with the decision tree approach.Potential application of the algorithms to domains related to petroleum engineering can help enhance understanding of the problems.

Stock Price Forecasting and Rule Extraction Based on L1-Orthogonal Regularized GRU Decision Tree Interpretation Model

Neural network is widely used in stock price forecasting,but it lacks interpretability because of its“black box”characteristics.In this paper,L1-orthogonal regularization method is used in the GRU model.A decision tree,GRU-DT,was conducted to represent the prediction process of a neural network,and some rule screening algorithms were proposed to find out significant rules in the prediction.In the empirical study,the data of 10 different industries in China’s CSI 300 were selected for stock price trend prediction,and extracted rules were compared and analyzed.And the method of technical index discretization was used to make rules easy for decision-making.Empirical results show that the AUC of the model is stable between 0.72 and 0.74,and the value of F1 and Accuracy are stable between 0.68 and 0.70,indicating that discretized technical indicators can predict the short-term trend of stock price effectively.And the fidelity of GRU-DT to the GRU model reaches 0.99.The prediction rules of different industries have some commonness and individuality.

Association Rules Mining Based on SVM and Its Application in Simulated Moving Bed PX Adsorption Process

In this paper, a novel data mining method is introduced to solve the multi-objective optimization problems of process industry. A hyperrectangle association rule mining （HARM） algorithm based on support vector machines （SVMs） is proposed. Hyperrectangles rules are constructed on the base of prototypes and support vectors （SVs） under some heuristic limitations. The proposed algorithm is applied to a simulated moving bed （SMB） paraxylene （PX） adsorption process. The relationships between the key process variables and some objective variables such as purity, recovery rate of PX are obtained. Using existing domain knowledge about PX adsorption process, most of the obtained association rules can be explained.

SVR-Miner:Mining Security Validation Rules and Detecting Violations in Large Software

For various reasons,many of the security programming rules applicable to specific software have not been recorded in official documents,and hence can hardly be employed by static analysis tools for detection.In this paper,we propose a new approach,named SVR-Miner(Security Validation Rules Miner),which uses frequent sequence mining technique [1-4] to automatically infer implicit security validation rules from large software code written in C programming language.Different from the past works in this area,SVR-Miner introduces three techniques which are sensitive thread,program slicing [5-7],and equivalent statements computing to improve the accuracy of rules.Experiments with the Linux Kernel demonstrate the effectiveness of our approach.With the ten given sensitive threads,SVR-Miner automatically generated 17 security validation rules and detected 8 violations,5 of which were published by Linux Kernel Organization before we detected them.We have reported the other three to the Linux Kernel Organization recently.

Research on A Web Intelligent Information Extraction Method

The paper introduce segmentation ideas in the pretreatment process of web page. By page segmentation technique to extract the accurate information in the extract region, the region was processed to extract according to the rules of ontology extraction, and ultimately get the information you need. Through experiments on two real datasets and compare with related work, experimental results show that this method can achieve good extraction results.

Enhanced Pattern Representation in Information Extraction

Traditional pattern representation in information extraction lack in the ability of representing domain-specific concepts and are therefore devoid of flexibility. To overcome these restrictions, an enhanced pattern representation is designed which includes ontological concepts, neighboring-tree structures and soft constraints. An information-(extraction) inference engine based on hypothesis-generation and conflict-resolution is implemented. The proposed technique is successfully applied to an information extraction system for Chinese-language query front-end of a job-recruitment search engine.

基于关联规则提取的外语翻译信息密度检测

为了提高外语翻译信息密度检测的有效性,提出基于关联规则提取的外语翻译信息密度检测方法。构建语料特征信息挖掘的二元模型,并采用关联规则提取方法计算关联规则分布特征参数,完成语料特征的提取。根据语料特征提取结果,计算信息密度检测中心点与信息增量特征的多维尺度信息。计算信息的平均可达密度,将其作为密度检测结果进行输出。实验结果表明,相较于传统对比方法,所提方法可以提高外语翻译信息密度检测精度,检测精度在97%左右。

高效液相色谱-质谱联用鉴定黄芩总苷元提取物中黄酮类成分

目的:使用高效液相色谱-四极杆-飞行时间串联质谱(HPLC-Q-TOF/MS)技术,分离鉴定黄芩总苷元提取物中的黄酮类成分,并总结其裂解规律。方法:采用岛津LX-20210330-01 C_(18)(250 mm×4.6 mm,5μm)色谱柱,以0.1%甲酸为流动相A,甲醇-乙腈(50∶50)为流动相B进行梯度洗脱,对黄芩总苷元提取物中的黄酮类成分进行分离,利用电喷雾离子化-四极杆-飞行时间串联质谱高分辨测定各成分母离子及其子离子的准确质量,结合电喷雾离子源正、负离子模式下的质谱信息和色谱保留时间进行结构鉴定。结果及结论:除了黄芩素和汉黄芩素两个已知成分外,从黄芩总苷元提取物中共鉴定出31种黄酮类成分,包括12种苷类与19种苷元,其中有3个化合物为本研究首次鉴定,并分析归纳了黄酮类成分在电喷雾离子源正、负离子模式下的质谱碎片裂解规律。

有限维空间下运动行为传感数据特征提取

维度的升高会加剧运动行为传感数据的复杂度,导致其分布特征空间被无限放大,因此提出基于有限维空间的运动行为传感数据特征提取方法。采用关联规则项挖掘分析方法计算数据模糊度,确定运动行为的有限空间区域。在有限维空间下,通过自适应寻优方法,计算传感数据的特征量化参数。检测运动行为传感数据的特征属性,计算数据分布融合映射输出结果,构建运动行为特征提取模型。实验结果表明,所提方法的运动数据空间聚类效果较好,能够把数据固定在有限维空间,数据特征提取精度始终保持在95%以上。

基于疾病网络的血友病并发症挖掘与关联规则分析

目的 基于疾病网络探讨血友病并发症的一般性规律,预测血友病患者可能发生的并发症。方法 从PubMed数据库中检索血友病相关文献,通过MetaMap工具从文献标题、摘要文本中抽取疾病实体,基于疾病对的共现关系构建疾病网络,分析疾病网络的整体特征、节点特征及结构特征等。对疾病实体网络进行关联分析,挖掘其关联规则,分析血友病并发症的一般性规律。采用链路预测算法预测血友病的潜在并发症。结果 血友病及其并发症构成的关联网络在网络结构上满足小世界网络特征和分布均匀的凝聚子群特征。凝聚子群分析结果显示,血友病并发症可分为遗传性疾病、血液系统疾病、传染性疾病和慢性疾病共4大类。关联规则分析发现133条置信度≥0.8的规则,链路预测进一步得到了许多有据可查的疾病对。结论 基于疾病网络进行血友病并发症关联分析和链路预测,可实现对血友病潜在并发症的有效预测,为血友病的临床诊疗提供决策支持。

基于词模式规则的轻量级日志模板提取方法

传统基于规则的日志解析方法针对每类日志需单独编写规则,且随着系统更新,出现新的日志模式时,需人工再次干预;基于深度学习的日志解析方法虽准确率高,但计算复杂度高。为解决日志解析方法人力成本和计算复杂度高的问题,文中提出一种基于词模式规则的轻量级日志模板提取方法,该方法由初始规则集生成、词模式规则应用、潜在错误样本发掘三个部分构成。首先,原始日志基于自适应随机抽样获取彼此间相似度较低的代表性日志;然后,基于专家反馈提取初始词模式规则集,在词模式规则应用模块对原始日志进行处理并提取日志模板;最后,在潜在错误样本发掘模块检查生成的日志模板聚类,发现潜在的错误分类样本并对其进行规则集更新。经过实验验证,在16个公开日志数据集上,文中方法的平均准确度达到97.8%,与基于深度学习的日志解析算法准确度基本持平;在计算效率方面,文中方法的单线程解析速度达到每秒20000条,且随着可用内核数量的增加,性能持续提升,满足系统日志的故障诊断和安全分析需求。

双行星排混合动力客车动态规划能量管理策略研究

为解决混合动力系统传统动态规划(Dynamic Programming,DP)策略成本函数值受罚函数影响较大、平衡电量调试时间较长的问题,针对某款双行星排混合动力客车提出了变惩罚系数DP(Variable Penalty Coefficient DP,DP-VAR)策略和边界迭代DP(Boundary Iterative DP,DP-ITE)策略。同时,还提出了基于XGBoost算法的动态规划策略模式切换规则提取方法,并对自适应等效燃油消耗最小策略(Adaptive Equivalent Fuel Consumption Minimization Strategy,A-ECMS)进行了改进,建立了具有模式切换规则的A-ECMS策略。基于C-WTVC工况的仿真结果表明:(1)相比于DP策略,DPVAR和DP-ITE策略的控制规则更为明确且百公里油耗分别降低了1.18%和4.31%;(2)DP-ITE策略在取消罚函数的前提下具有终止SOC绝对收敛性;(3)优化后的A-ECMS策略较优化前的百公里油耗降低了1.17%。

利用集成剪枝和多目标优化算法的随机森林可解释增强模型

随机森林模型是广泛应用于各个领域的经典黑盒模型,而黑盒模型的结构特征导致模型可解释性弱,需要借助可解释技术优化随机森林的可解释性,从而促进其在可靠性要求较高场景的应用与发展。研究构建了基于集成剪枝和多目标优化算法的规则提取模型,集成剪枝在解决树模型规则提取易陷入局部最优的问题上具有代表性,多目标优化在解决规则准确性和可解释性的平衡问题上有多个领域的应用。模型验证结果表明,所构建模型能够在不降低准确性的前提下优化模型的可解释性。本研究首次将集成剪枝技术与多目标优化算法相融合,增强了随机森林的可解释性,有助于推动该模型在可解释性要求较高领域的决策应用。