IT-Application Behaviour Analysis: Predicting Critical System States on OpenStack using Monitoring Performance Data and Log Files

In order to solove the aero-engine maintenance level and costs optimization, it analyzed the practical monitoring performance data of aero-engine and proposed a method to mine maintenance decision-making rules by the rough sets theory. In the process of data mining, a self-organizing maps neural network(SOM-NN) was used to the continuous data discretization and obtained better effect. CF6-80C2A5 is illustrated to mine the relevance and decision-making rules between maintenance levels and condition parameters based on its monitoring performance data. The case study indicates the acquired maintenance decision-making rules are possessed of obvious practical value for airlines engineers by the method proposed herein.

This paper addresses the question of whether administrative data sources, such as performance monitoring data, can be used for program evaluation purposes. It argues that under certain circumstances, such data can be used. In particular, program performance data that are routinely gathered and monitored by administrators of many workforce development programs meet these circumstances. The paper goes on to demonstrate the point by using administrative data from the state of Washington to examine the net impact on earnings and employment of services provided to adults under the Workforce Investment Act (WIA). Because of a lack of consensus about appropriate net impact estimators, the strategy of this paper is to examine the sensitivity of the results to various estimation techniques. The paper describes the various estimation techniques, and it summarizes the net impact estimates that are generated for the State of Washington. For the most part, the results are fairly stable across the techniques, which the paper argues adds a degree of confidence in them. The final section of the paper offers guidance to policymakers and program administrators who may not be familiar with the technical details of various analytical approaches about how empirical results that may appear to be complex or unstable can be used for program improvement.

Today, it is almost impossible to implement teaching processes without using information and communication technologies (ICT), especially in higher education. Education institutions often use learning management systems (LMS), such as Moodle, Edmodo, Canvas, Schoology, Blackboard Learn, and others. When accessing these systems with their personal account, each student's activity is recorded in a log file. Besides analyzing the raw data from log files directly, there is an option to use Moodle plugins that provide learning analytics and enable the faster analysis of students' behavior on LMS. In this paper, some of these plugins are presented. However, this paper is focused on analyzing the log files of a course implemented on the LMS Moodle at the Faculty of Organization and Informatics at the University of Zagreb. The results of the students' behavior, based on logs in Moodle, will be interpreted in terms of student success.

Resource-intensive applications such as scientific applications require the architecture or system on which they execute to display a very high level of dependability to reduce the impact of faults. Typically, the state of the underlying system is captured through messages that are recorded in a log file, which has been proven useful to system administrators in understanding the root-causes of system failures (and for their subsequent debugging). However, the time window between when the first error message is detected in the log file and time of the ensuing failure may not be large enough to allow the administrators to save the state of the running application, which will result in lost execution time. We thus address this fundamental question: Is it possible to extend this time window? The answer is positive: We show that, by using (i) resource usage logs to track anomalous resource usage and (ii) error logs to identify root-causes of system failures, it is possible to increase the time window, on average, by 50 minutes. These files were those obtained for the Ranger Supercomputer from TACC. We achieve this by applying anomaly detection techniques on resource usage data and conducting a root-cause analysis on error log files.

This paper mainly analyzes the fracturing effect of horizontal wells in the development process of horizontal wells in low permeability A reservoir. On the basis of tracer monitoring, microseismic monitoring and production performance data, it is considered that fracturing breakthrough can be basically realized with the current fracturing scale of horizontal wells, but the actual breakthrough effect of communication layer depends on synchronous water injection to maintain formation energy. At the same time, the large-scale fracturing of horizontal wells can form a network fracture body in the formation, which is conducive to the production of oil layers near the well. Based on this conclusion, the next step is to formulate water injection policy for horizontal well development, so as to provide basis for future development research and water injection adjustment of horizontal wells in the same type of oilfield.

Web mining aims to discover useful information or knowledge from the web hyperlink structure, page content and usage log. The “user-centered” philosophy of this tool is in perfect harmony with the concepts of modern marketing, ergonomics, and learning management. This new approach, as opposed to the traditional “page-centered” philosophy, puts the users' goals and intentions to the centre, and designs the services of the system accordingly. About 60 learners of technical teacher training took part simultaneously in processing the Educational technology and multimedia course, all their activities performed in Moodle learning environment were registered in a log file by the server. The processing of this log file was performed by the SPSS Web Mining for Clementine program. Here we are going to present the first results exposed by quality assurance in connection with the students' learning activity, the structure of the syllabus as well as the navigational opportunities. By analysis of student behavior we receive some beneficial information for course development and learning management which are applied in design of adaptive online learning in engineering education.

Internal user threat detection is an important research problem in the field of system security. Recently, the analysis of abnormal behaviors of users is divided into supervised learning method (SID) and unsupervised learning method (AD). However, supervised learning method relies on domain knowledge and user background, which means it cannot detect previously unknown attacks and is not suitable for multi-detection domain scenarios. Most existing AD methods use the clustering algorithm directly. But for threat detection on internal users’ behavior, mostly for high-dimensional cross-domain log files, as far as we know, there are few methods of multi-domain audit log data with effective feature extraction. An effective feature extraction method which can not only reduce testing cost greatly, but also detect the abnormal behavior of users more accurately. We propose a new unsupervised log abnormal behavior detection method which is based on the denoising autoencoders to encode the user log file, and adopts the integrated method to detect the abnormal data after encoding. Compared with the traditional detection method, it can analyze the abnormal information in the user behavior more effectively, thus playing a preventive role against internal threats. In addition, the method is completely data driven and does not rely on relevant domain knowledge and user’s background attributes. Experimental results verify the effectiveness of the integrated anomaly detection method under the multi-domain detection scenario of user log files.

Computer forensics aimed at determining the relevant causes and effects between the present state of computer system and malicious operation through the measures which could be adopted on the court. This paper proposes an intrusion forensics worked on a WEB server. The forensics system monitor the access to the log files and combine it with the timestamp and other clues in the log file, thus comprehensive operation facts are formed and represented by XML. Through analyzing with a decision tree, intrusion behavior evidence can be abstracted. Especially when a hacker tried to wipe his trace, the system can detect it effectively.

Modern mass customization production allows user interaction activities to be distributed in the full product life cycle via multiple information systems. Investigating user behaviors across the boundaries of different domains helps to deeply integrate isolated fragmental profiles into a comprehensive one, and therefore can provide multi-dimensional, high-quality and valuable services. However, traditional user behavior analysis models are based on individual user profile information derived from separate domains, such as requirement analysis, design, supply chain, logistics, marketing, etc., which have not considered the whole complexity of mass customization manufacturing. In this paper, we introduce the concept of multi-dimensional semantic activity space, where user behavior features are merged and represented as combined vectors. User behavior patterns are discovered by mining action data extracted from log files in different subsystems in the corresponding domains. We also identify distinct categories of user behaviors in various modules and subsystems in the context of an intelligent manufacturing environment. Experiment results show a strong indication that the proposed approach can be applied to reveal variations in typical behavioral aspects of cross-domain participants, in terms of patterns in resource access, operation tasks, performance assessment, etc.

Introduction. This paper discusses causes behind failed searches from the searchers’ perspective and examines associated emotions. Method. We conducted an online study with real users in their natural settings. Participants were asked to use a digital library and run one specific and one exploratory search task, using their own topics. They also answered pre- and post-questionnaires for both task types. Analysis. Three types of analysis were conducted: (i) one descriptive on answers in questionnaires, (ii) one textual on users’ failure reports to identify the causes and elicit associated emotions and (iii) one behavioural analysis of the interactions recorded in log files. Results. Users identified the lack of coverage and poor usability as main causes behind failed searches. We examined search behaviour along with declared perceptions of the causes behind failures and realised that the digital library functionalities were not fully used. Poor awareness of the digital library functionalities could be an unreported cause for search failures. In general, users had a positive attitude toward the digital library, expressing trust, joy, and anticipation. Anger and sadness were linked specifically to failed searches. Conclusions. There persists a need to improve the digital library systems and simplify their interfaces. The emotional effect is a significant factor that need to be considered in the user experience studies.

This paper has proposed a shopping assistance service. We provide consumer-friendly services which based on personal behavior log data. The consumer-friendly services are information service which display through a monitor, robot, sound and music that can attract new consumer. In this paper, we study personal behavior from a log file that has been recorded by RFID and camera sensors in the ubiquitous environment. Utilizing ubiquitous sensors, we apply three types of observation scopes to model each user's preference. They are microscopic, mezzoscopic, and macroscopic observation scopes. We have proposed the behavior analysis, algorithms and collaborative filtering to retrieve information. Finally, we will also propose a ubiquitous shop space experiment.

A 3D visual analysis tool has been developed to add value to the SANDF's growing Ground Based Air Defence (GBAD) System of Systems simulation capability. A time based XML interface between the simulation and analysis tool, via a TCP connection or a log file, allows individual simulation objects to be wholly updated or partially modified. Live pause and review of the simulation action is supported by employing data key frames and compressed XML for enhanced performance. An innovative configurable filter tree allows visual clutter to be reduced as required and an open source scene graph (OpenSceneGraph) manages the 3D scene representation and rendering.A visualisation capability is developed for the effective presentation of the dynamic air defence system behaviour, system state transitions and inter-system communication. The visual analysis tool has successfully been applied in support of system performance experiments, tactical doctrine development and simulation support during training and live field exercises. The 3D visualisation resulted in improved situational awareness during experiment analysis, in increased involvement of the SANDF in experiment analysis and in improved credibility of analysis results presented during live or after action visual feedback sessions.

Laws are derived for stationary states in self-organized critical systems with extremal dynamics. In a class of models, the exponent $\ensuremath{\eta}$ for the survival of activity is zero in all dimensions. Scaling relations for other critical exponents are found. Numerical simulations of Sneppen's interface depinning model, Zaitsev's low temperature creep model, and the Bak-Sneppen evolution model agree with these predictions.

The current paper presents the results of the evaluation of a novel human machine interface for supervision and guidance of multiple highly automated unmanned aircraft. Accidents of unmanned aircraft are often related to deficiencies in the design of the human machine interface. Many of these deficiencies are a consequence of insufficient consideration of human factors engineering principles in the early stages of system development. Therefore, in the design of the current human machine interface, empirically validated human factors methods were applied in order to optimally support a remote pilot in supervising and guiding multiple highly automated unmanned aircraft. An information management concept was developed that allows for an increase in the number of UAS that are being supervised without considerably increasing display complexity. In order to evaluate the human machine interface, two within-subj ect simulation studies were conducted. In both studies, pilots were asked to supervise and guide a varying number of UAS in several experimental scenarios using the new human machine interface. Following the experimental scenarios, a structured interview regarding the design of the human machine interface was conducted. The objective of the first study was to gather initial data on the general usability of the interface. The aim of the second simulation study was to investigate how increasing the number of UAS from one up to five affects the time that a pilot needs to acquire and change specific system parameters of any UAS under supervision. Further, the impact of increasing the number of UAS on pilot workload and situation awareness was assessed. The results of the interviews suggest that the design concept of the human machine interface is feasible and easy to understand. The information management concept was described as reasonable and logical. Important information was accessible quickly and (critical) system states could easily be identified. No significant effects of increasing the number of UASs to be supervised on reaction, workload or situation awareness were observed. The findings suggest that the information management concept is reasonable and usable.

The role of disorder and stress concentration in nonconservative fault systems