Classification Of Logs Research Articles

Design and implementation of real-time log analysis based on ELK

With the rapid development of the IT industry, the integrated solution of real-time log analysis has become a member of the new generation of IT operation and maintenance structure. The traditional manual log analysis system is no longer suitable for the rapid development of the operation and maintenance system. The log analysis software commonly used in enterprises includes Elasticsearch, Logstash, and Kibana systems. The advantage is that it is free and open source. "Design and implementation of real-time log analysis based on ELK" was born for system maintenance and developers to understand server software and hardware information through logs, and check the cumbersome operations in the configuration process. Because frequent analysis of the log can understand the load of the server, performance security, so as to take timely measures to correct errors. If you manage dozens or hundreds of servers, there is no centralized log management system to collect and summarize the logs on all servers, which causes great difficulties for operation and maintenance managers. "Design and implementation of real-time log analysis based on ELK" uses Elasticsearch, Logstash, and Kibana to achieve unified collection and classification of server logs, which can solve the defects of traditional log analysis platforms, improve the efficiency of server application positioning problems, and help enterprises solve data center problems running problems of related applications.

Enhancing the automatic facies classification of Brazilian presalt acoustic image logs with SwinV2-Unet: Leveraging transfer learning and confident learning

Facies classification of image logs plays a vital role in reservoir characterization, especially in the heterogeneous and anisotropic carbonate formations of the Brazilian presalt region. Although manual classification remains the industry standard for handling the complexity and diversity of image logs, it has the notable disadvantages of being time consuming, labor intensive, subjective, and nonrepeatable. Recent advancements in machine learning offer promising solutions for automation and acceleration. However, previous attempts to train deep neural networks for facies identification have struggled to generalize to new data due to insufficient labeled data and the inherent intricacy of image logs. In addition, human errors in manual labels further hinder the performance of trained models. To overcome these challenges, we develop adopting the state-of-the-art SwinV2-Unet to provide depthwise facies classification for Brazilian presalt acoustic image logs. The training process incorporates transfer learning to mitigate overfitting and confident learning to address label errors. Through a k-fold cross-validation experiment, with each fold spanning more than 350 m, we achieve an impressive macro F1 score of 0.90 for out-of-sample predictions. This significantly surpasses the previous model modified from the widely recognized U-Net, which provides a macro F1 score of 0.68. These findings highlight the effectiveness of the used enhancements, including the adoption of an improved neural network and an enhanced training strategy. Moreover, our SwinV2-Unet enables a highly efficient and accurate facies analysis of the complex yet informative image logs, significantly advancing our understanding of hydrocarbon reservoirs, saving human effort, and improving productivity.

Classification of e-commerce financial transaction logs using machine learning approach

E-Commerce becomes inevitable in the current world. Especially in this pandemic period, almost activities have been carried out through digital mode to serve the customers with more safety. The financial transactions are much secured in the e-commerce payment. Still, many intruders are making these transactions into fail and hacking the customers’ information when the financial transactions are carried out. Along with, sometimes the transactions may get failed due to the network issues. Hence, the E-Commerce organizations are maintaining the transaction logs to check and take necessary action over the failed transactions. Out of huge transaction logs, identifying the suspicious failed transactions in manual method is not encouraged. Lot of technologies have come to support the detection of suspicious failed transactions such as Artificial Neural Network, Machine Learning, Deep Learning and other statistical methods. As the above methods are proved as more suitable for detecting and classifying the failed transaction logs, still the accuracy of classification has not been achieved much, which motivated to propose the machine learning based classification of E-Commerce Financial Transaction Logs.

Research on Knowledge Graph Model for Cybersecurity Logs Based on Ontology and Classified Protection

In order to audit and analyse cybersecurity events from massive logs, the knowledge graph model for cybersecurity logs is proposed. The data layer and the pattern layer of the knowledge graph model for cybersecurity logs is obtained based on ontology and classified protection. Entity classification of cybersecurity logs is extracted by using classified protection standards in the data layer of the knowledge graph. The relationship and attribute of ontology from cybersecurity logs is defined. The pattern layer of the knowledge graph model for cybersecurity logs is defined, classified protection data is integrated in entity alignment, and classified protection information gain is used in ontology construction. The structure of the knowledge graph for cybersecurity logs is given. So that the efficient association and deep mining analysis of cybersecurity logs are realized, and it can be directly analysed and processed on the data without the need for precise modelling of the problem. The efficiency of logs analysis is improved. The model has heuristic characteristics and generalization abstract ability, and the model is suitable for big data analysis of large-scale cybersecurity logs.

Cost-based text understanding to improve maintenance knowledge intelligence in manufacturing enterprises

Improving maintenance knowledge intelligence using text data has not been largely explored in the literature of production and engineering management. The state-of-the-art approaches and solutions mainly focus on either clustering and classification of maintenance logs, or extracting additional (meta-)data e.g. failure time data from maintenance text reports, operators’ workbooks and digital logbook. Knowledge Discovery from Text (KDT) enables finding undetected causalities, hidden patterns, frequencies, associative relations, and sentiments in maintenance text repositories. Applying KDT may enhance understanding the content of text data syntactically and semantically. However, advanced KDT approaches do not significantly provide meaningful and explainable outcomes, due to certain barriers in manufacturing enterprises, namely availability and quality of (longitudinal) maintenance text data.To overcome these barriers in real world industrial maintenance, generate added value in industrial maintenance, and lay the ground for autonomous maintenance decision-support in the context of Industry 4.0, the first step is to adopt KDT methods and accordingly provide maintenance-specific solutions considering practical challenges and possibilities.This paper discusses the lack of understanding maintenance text data and examines its effect on maintenance knowledge intelligence in manufacturing enterprises. A compositional framework for text understanding (TextPlan) is introduced. TextPlan explores quantification of text data in both syntax and semantic levels, i.e. how to vectorize an annotated maintenance report into numeric values, which represent cost data, hidden associations and sentiments. A prominent feature of TextPlan is cost-based text analysis, which decomposes a maintenance text report into separate cost items, and then (re-)composes the findings to estimate the total maintenance cost associated with the given report. Finally yet importantly, TextPlan consolidates the findings into a Text Understanding Map for assisting maintenance planner, based on three proposed measures of text comprehension, namely Association Measuring Index (AMI), Opinion Index (OI) and Cost Vector (CV).

Hierarchical Classification and Visualization Method of Social Simulation Logs Reflecting Multiple Analytical Interests

Hierarchical Classification and Visualization Method of Social Simulation Logs Reflecting Multiple Analytical Interests

Classification of Firewall Logs Using Supervised Machine Learning Algorithms

Classification of Firewall Logs Using Supervised Machine Learning Algorithms

Robust Library Building for Autonomous Classification of Downhole Geophysical Logs Using Gaussian Processes

Machine learning techniques such as Gaussian Processes can be used to identify stratigraphically important features in geophysical logs. The marker shales in the banded iron formation hosted iron ore deposits of the Hamersley Ranges, Western Australia, form distinctive signatures in the natural gamma logs. The identification of these marker shales is important for stratigraphic identification of unit boundaries for the geological modelling of the deposit. Machine learning techniques each have different unique properties that will impact the results. For Gaussian Processes (GPs), the output values are inclined towards the mean value, particularly when there is not sufficient information in the library. The impact that these inclinations have on the classification can vary depending on the parameter values selected by the user. Therefore, when applying machine learning techniques, care must be taken to fit the technique to the problem correctly. This study focuses on optimising the settings and choices for training a GPs system to identify a specific marker shale. We show that the final results converge even when different, but equally valid starting libraries are used for the training. To analyse the impact on feature identification, GP models were trained so that the output was inclined towards a positive, neutral or negative output. For this type of classification, the best results were when the pull was towards a negative output. We also show that the GP output can be adjusted by using a standard deviation coefficient that changes the balance between certainty and accuracy in the results.

A comparative study of 3D FZI and electrofacies modeling using seismic attribute analysis and neural network technique: A case study of Cheshmeh-Khosh Oil field in Iran

Abstract Electrofacies are used to determine reservoir rock properties, especially permeability, to simulate fluid flow in porous media. These are determined based on classification of similar logs among different groups of logging data. Data classification is accomplished by different statistical analysis such as principal component analysis, cluster analysis and differential analysis. The aim of this study is to predict 3D FZI (flow zone index) and Electrofacies (EFACT) volumes from a large volume of 3D seismic data. This study is divided into two parts. In the first part of the study, in order to make the EFACT model, nuclear magnetic resonance (NMR) log parameters were employed for developing an Electrofacies diagram based on pore size distribution and porosity variations. Then, a graph-based clustering method, known as multi resolution graph-based clustering (MRGC), was employed to classify and obtain the optimum number of Electrofacies. Seismic attribute analysis was then applied to model each relaxation group in order to build the initial 3D model which was used to reach the final model by applying Probabilistic Neural Network (PNN). In the second part of the study, the FZI 3D model was created by multi attributes technique. Then, this model was improved by three different artificial intelligence systems including PNN, multilayer feed-forward network (MLFN) and radial basis function network (RBFN). Finally, models of FZI and EFACT were compared. Results obtained from this study revealed that the two models are in good agreement and PNN method is successful in modeling FZI and EFACT from 3D seismic data for which no Stoneley data or NMR log data are available. Moreover, they may be used to detect hydrocarbon-bearing zones and locate the exact place for producing wells for the future development plans. In addition, the result provides a geologically realistic spatial FZI and reservoir facies distribution which helps to understand the subsurface reservoirs heterogeneities in the study area.

Query Topic Classification and Sociology of Web Query Logs

In the paper, the objects, tasks, and a general procedure of the sociological analysis of Web search engine query logs are described and illustrated by a methodologically complete study of the cross-nation search image changes based on two-year spaced query logs of the national search audience.

Automatic classification of communication logs into implementation stages via text analysis.

BackgroundTo improve the quality, quantity, and speed of implementation, careful monitoring of the implementation process is required. However, some health organizations have such limited capacity to collect, organize, and synthesize information relevant to its decision to implement an evidence-based program, the preparation steps necessary for successful program adoption, the fidelity of program delivery, and the sustainment of this program over time. When a large health system implements an evidence-based program across multiple sites, a trained intermediary or broker may provide such monitoring and feedback, but this task is labor intensive and not easily scaled up for large numbers of sites.We present a novel approach to producing an automated system of monitoring implementation stage entrances and exits based on a computational analysis of communication log notes generated by implementation brokers. Potentially discriminating keywords are identified using the definitions of the stages and experts’ coding of a portion of the log notes. A machine learning algorithm produces a decision rule to classify remaining, unclassified log notes.ResultsWe applied this procedure to log notes in the implementation trial of multidimensional treatment foster care in the California 40-county implementation trial (CAL-40) project, using the stages of implementation completion (SIC) measure. We found that a semi-supervised non-negative matrix factorization method accurately identified most stage transitions. Another computational model was built for determining the start and the end of each stage.ConclusionsThis automated system demonstrated feasibility in this proof of concept challenge. We provide suggestions on how such a system can be used to improve the speed, quality, quantity, and sustainment of implementation. The innovative methods presented here are not intended to replace the expertise and judgement of an expert rater already in place. Rather, these can be used when human monitoring and feedback is too expensive to use or maintain. These methods rely on digitized text that already exists or can be collected with minimal to no intrusiveness and can signal when additional attention or remediation is required during implementation. Thus, resources can be allocated according to need rather than universally applied, or worse, not applied at all due to their cost.

DETERMINAÇÃO DO RENDIMENTO DE MATÉRIA-PRIMA DE Eucalyptus benthamii Maiden & Cambage POR MEIO DE DIFERENTES MÉTODOS DE DESDOBRO

RESUMOO objetivo deste trabalho foi determinar o rendimento da matéria-prima de Eucalyptus benthamii, por meio dos desdobros tangencial e radial, além da classificação das toras quanto à presença das rachaduras de topo. A classificação das toras quanto às rachaduras foi realizada por meio da análise visual, em que era aplicado um conceito conforme a intensidade do defeito. Posteriormente, as toras foram separadas em duas classes diamétricas (20 a 25 e 25,1 a 30 cm) e processadas em três sistemas de desdobro (tangencial em bloco, radial e tangencial em bloco com face curva), sendo cada tratamento composto por três repetições com 10 toras cada um. Na análise estatística dos dados de rendimento, foram aplicados o teste de Kolmogorov Smirnov, a Análise da Variância e o teste de Tukey. Nos resultados, foi observado que a maioria das toras da classe de menor diâmetro apresentou intensidade mediana, enquanto na maior classe diamétrica a intensidade foi baixa. Os rendimentos médios líquidos encontrados foram de 43,5% e 39,2% nas classes inferior e superior, respectivamente, os quais foram influenciados pelas rachaduras das peças e pelos destopos. Observou-se que o desdobro tangencial em bloco e em face curva foram os métodos que apresentaram maior aproveitamento em madeira serrada na classe de 20 a 25 cm, enquanto na classe de 25,1 a 30 cm o sistema tangencial em bloco e radial foi equivalente.

Automatic classification of communication logs into implementation stages via text analysis

Automatic classification of communication logs into implementation stages via text analysis

小学校低学年の体育授業における学習集団の形成過程に関する事例研究

The purpose of this case study was to clarify how pupils improve the quality of a learning group, gakushu shudan, while learning the contents of a physical education class, and to examine how a teacher should instruct pupils in order to improve the quality of learning. The gakushu shudan is a concept that aims to promote integral acquisition of subject contents and subsequent character formation in a class.The class study was conducted using short sequences of movement to accompany a story and a song, shudan ohanashi mat for second grade elementary school children. In the shudan ohanashi mat class, two learning activities were conducted. First, pupils discussed the constitution of the short movement sequences. Secondly, pupils synchronized the movement with another member of the group. We then extracted two groups for analysis in detail: one that showed a marked difference in motor skills among the group members and another that showed few differences in motor skills. We then conducted three types of analysis. First, we examined the general content learned by pupils by classification of the learning logs. Secondly, we examined the formative evaluation of the children's group activity through analysis of a “making friends” questionnaire completed by the pupils. Thirdly, we analyzed the utterance of these pupils through “discourse analysis” of characteristic classes.The “discourse analysis” revealed that differences not only in motor skills but also the social background of the pupils affected the formation of the learning group. In a group of 7, D (a girl with low motor skills) began to learn positively with other members of the group when she acquired technical recognition. In a group of 8, a change in the manner of H (a girl with high motor skills) generated good communication between the group members.We conclude that in the early period of formation of a learning group, the following three factors appear to have a positive influence.1) Teacher's guidance in allowing pupils to share means of recognition.2) Teacher's guidance in allowing pupils to interact positively.3) Teacher's approach to group organization by selecting the pupils of democratic nature as leaders.

Rule-driven defect detection in CT images of hardwood logs

This paper deals with automated detection and identification of internal defects in hardwood logs using computed tomography (CT) images. We have developed a system that employs artificial neural networks (ANNs) to perform tentative classification of logs on a pixel-by-pixel basis. This approach achieves a high level of classification accuracy for several hardwood species (northern red oak, Quercus rubra, L., water oak, Q. nigra, L., yellow poplar, Liriodendron tulipifera, L., and black cherry, Prunus serotina, Ehrh.), and three common defect types (knots, splits, and decay). Although the results are very satisfactory statistically, a subjective examination reveals situations that could be refined in a subsequent post-processing step. We are currently developing a rule-based, contextual approach to region refinement that augments the initial emphasis on local information. The resulting rules are domain dependent, utilizing information that depends on region shape and type of defect. For example splits tend to be long and narrow, and this knowledge can be used to merge smaller, disjoint regions that have tentatively been labeled as splits. Similarly, image regions that represent knots, decay, and clear wood can be refined by removing small, spurious regions and by smoothing the boundaries of these regions. Mathematical morphology operators can be used for most of these tasks. This paper provides details concerning the domain-dependent rules by which morphology operators are chosen, and for merging results from different operations.

Using forest stand reconstructions to assess the role of structural continuity for late-successional species

Historical reconstructions of past forest dynamics and stand structures have been used to establish reference conditions for managing present forest ecosystems. In this study we (1) developed and combined a suite of stand reconstruction techniques to describe past stand characteristics, and (2) applied these stand histories to evaluate the relationship between wood-decay fungi and forest continuity. Ten previous selectively logged stands of Norway spruce ( Picea abies L. Karst.), in the middle boreal zone of southeastern Norway, were studied. We reconstructed stand structures during the 20th century using tree-ring series, growth patterns, age structures, and decay classification and datings of stumps and logs. All stands were selectively logged between 1890 and 1965, with a mean logging interval of 25 years. Harvested volumes (1900–1965) constituted 25–99% of present standing volumes and present volumes were 2.6–21 (median 4) times higher than the lowest estimated historic volumes. Dead wood was categorized into eight decay classes, where one is recently fallen, and eight is almost completely decayed. Six fungus species, assumed to indicate dead-wood continuity, were found on logs in decay classes 2–4, all of which were estimated to be <30 years old. Logs in decay classes 1–4 constituted 85% of logs ≥20 cm. Expectedly, fungus abundance increased linearly with increasing number of available logs, but we failed to find a positive correlation between fungi abundance and number of old logs present (decay classes 5–8), when the effect of younger logs (2–4) was accounted for. This finding, together with the stand histories, does not lend support to the hypothesis that a continuous supply of dead wood, at the scale of forest stands, is crucial for the occurrence of the surveyed wood-decay fungi. We propose forest stand reconstructions to hold promise as a tool to assess the role of structural continuity for the occurrence of late-successional and old-growth species.

Visualizing Heterogeneous Clastic Reservoirs: Price Formation (Early Mississippian) Oil Fields in West Virginia: ABSTRACT

A procedure has been developed using public-domain and published software for creating 3-D models of facies in structurally-deformed elastic reservoirs. The procedure was tested successfully on two oil fields in Early Mississippian sandstone reservoirs in central West Virginia. The procedure has four stages: removal of structural deformation; definition of electrofacies from digitized electric logs and cores; classification of additional logs; and 3-D kriging. Reconstruction of original bedding was needed before variography and kriging, but no datum was available: the reservoir sandstone lies just below an angular unconformity, and many wells were not drilled far enough below the reservoir to penetrate a potential datum. In a novel application of automated stratigraphic correlation, a published algorithm was used on digitized gamma-ray logs to find the relative vertical shift between pairs of wells giving the highest cross-correlation. Multidimensional scaling of a matrix of shifts yielded a vector of values necessary for restoring the relative elevation of each well. Cluster analysis of gamma-ray and density log responses from cored wells defined four groups matched through core descriptions with environments of deposition. Discriminant functions calculated for these groups were used to classify log responses from uncored wells. Kriging of electrofacies followed conventional variography. Results aremore » displayed as cross-sections and maps.« less