Show-and-Tell; an interactive programming system for image processing.

Includes bibliographical references.

DMS : Diagnostic Modeling System.

Includes bibliographical references (p. 52-53).

A guide to major job accounting systems : the Logger system of the UNIVAC 1100 series operating system /

Includes bibliographical references.

Fitness assessment of document model

Document classification is a supervised machine learning process, where predefined category labels are assigned to documents based on the hypothesis derived from training set of labelled documents. Documents cannot be directly interpreted by a computer system unless they have been modelled as a collection of computable features. Rogati and Yang [M. Rogati and Y. Yang, Resource selection for domain-specific cross-lingual IR, in SIGIR 2004: Proceedings of the 27th annual international conference on Research and Development in Information Retrieval, ACM Press, Sheffied: United Kingdom, pp. 154-161.] pointed out that the effectiveness of document classification system may vary in different domains. This implies that the quality of document model contributes to the effectiveness of document classification. Conventionally, model evaluation is accomplished by comparing the effectiveness scores of classifiers on model candidates. However, this kind of evaluation methods may encounter either under-fitting or over-fitting problems, because the effectiveness scores are restricted by the learning capacities of classifiers. We propose a model fitness evaluation method to determine whether a model is sufficient to distinguish positive and negative instances while still competent to provide satisfactory effectiveness with a small feature subset. Our experiments demonstrated how the fitness of models are assessed. The results of our work contribute to the researches of feature selection, dimensionality reduction and document classification.

An investigation into inductive parameter learning in complex hierarchical knowledge structures representing clinical expertise

This dissertation investigates the very important and current problem of modelling human expertise. This is an apparent issue in any computer system emulating human decision making. It is prominent in Clinical Decision Support Systems (CDSS) due to the complexity of the induction process and the vast number of parameters in most cases. Other issues such as human error and missing or incomplete data present further challenges. In this thesis, the Galatean Risk Screening Tool (GRiST) is used as an example of modelling clinical expertise and parameter elicitation. The tool is a mental health clinical record management system with a top layer of decision support capabilities. It is currently being deployed by several NHS mental health trusts across the UK. The aim of the research is to investigate the problem of parameter elicitation by inducing them from real clinical data rather than from the human experts who provided the decision model. The induced parameters provide an insight into both the data relationships and how experts make decisions themselves. The outcomes help further understand human decision making and, in particular, help GRiST provide more accurate emulations of risk judgements. Although the algorithms and methods presented in this dissertation are applied to GRiST, they can be adopted for other human knowledge engineering domains.

Computer Processing of Medieval Slavic Sources in the Institute of Literature at BAS Repertorium Project (1994–2004)

Mixed-content miscellanies (very frequent in the Byzantine and mediaeval Slavic written heritage) are usually defined as collections of works with non-occupational, non-liturgical application, and texts in them are selected and arranged according to no identifiable principle. It is a “readable” type of miscellanies which were compiled mainly on the basis of the cognitive interests of compilers and readers. Just like the occupational ones, they also appeared to satisfy public needs but were intended for individual usage. My textological comparison had shown that mixed- content miscellanies often showed evidence of a stable content – some of them include the same constituent works in the same order, regardless that the manuscripts had no obvious genetic relationship. These correspondences were sufficiently numerous and distinctive that they could not be merely fortuitous, and the only sensible interpretation was that even when the operative organizational principle was not based on independently identifiable criteria, such as the church calendar, liturgical function, or thematic considerations, mixed-content miscellanies (or, at least, portions of their contents) nonetheless fell into types. In this respect, the apparent free selection and arrangement of texts in mixed-content miscellanies turns out to be illusory. The problem was – as the corpus of manuscripts that I and my colleagues needed to examine grew – our ability to keep track of the structure of each one, and to identify structural correspondences among manuscripts within the corpus, diminished. So, at the end of 1993 I addressed a letter to Prof. David Birnbaum (University of Pittsburgh, PA) with a request to help me to solve the problem. He and my colleague Andrey Boyadzhiev (Sofia University) pointed out to me that computers are well suited to recording, processing, and analyzing large amounts of data, and to identifying patterns within the data, and their proposal was that we try to develop a computer system for description of manuscripts, for their analysis and of course, for searching the data. Our collaboration in this project is now ten years old, and our talk today presents an overview of that collaboration.

Multi-Agent Security System based on Neural Network Model of User's Behavior

It is proposed an agent approach for creation of intelligent intrusion detection system. The system allows detecting known type of attacks and anomalies in user activity and computer system behavior. The system includes different types of intelligent agents. The most important one is user agent based on neural network model of user behavior. Proposed approach is verified by experiments in real Intranet of Institute of Physics and Technologies of National Technical University of Ukraine "Kiev Polytechnic Institute”.

Computer Mathematics with MuPAD in School Years 6. To 9

Report published in the Proceedings of the National Conference on "Education and Research in the Information Society", Plovdiv, May, 2014

Actionable pattern analysis in computing and business intelligence systems

With advances in science and technology, computing and business intelligence (BI) systems are steadily becoming more complex with an increasing variety of heterogeneous software and hardware components. They are thus becoming progressively more difficult to monitor, manage and maintain. Traditional approaches to system management have largely relied on domain experts through a knowledge acquisition process that translates domain knowledge into operating rules and policies. It is widely acknowledged as a cumbersome, labor intensive, and error prone process, besides being difficult to keep up with the rapidly changing environments. In addition, many traditional business systems deliver primarily pre-defined historic metrics for a long-term strategic or mid-term tactical analysis, and lack the necessary flexibility to support evolving metrics or data collection for real-time operational analysis. There is thus a pressing need for automatic and efficient approaches to monitor and manage complex computing and BI systems. To realize the goal of autonomic management and enable self-management capabilities, we propose to mine system historical log data generated by computing and BI systems, and automatically extract actionable patterns from this data. This dissertation focuses on the development of different data mining techniques to extract actionable patterns from various types of log data in computing and BI systems. Four key problems—Log data categorization and event summarization, Leading indicator identification , Pattern prioritization by exploring the link structures , and Tensor model for three-way log data are studied. Case studies and comprehensive experiments on real application scenarios and datasets are conducted to show the effectiveness of our proposed approaches.

Temporal Mining for Distributed Systems

Many systems and applications are continuously producing events. These events are used to record the status of the system and trace the behaviors of the systems. By examining these events, system administrators can check the potential problems of these systems. If the temporal dynamics of the systems are further investigated, the underlying patterns can be discovered. The uncovered knowledge can be leveraged to predict the future system behaviors or to mitigate the potential risks of the systems. Moreover, the system administrators can utilize the temporal patterns to set up event management rules to make the system more intelligent. With the popularity of data mining techniques in recent years, these events grad- ually become more and more useful. Despite the recent advances of the data mining techniques, the application to system event mining is still in a rudimentary stage. Most of works are still focusing on episodes mining or frequent pattern discovering. These methods are unable to provide a brief yet comprehensible summary to reveal the valuable information from the high level perspective. Moreover, these methods provide little actionable knowledge to help the system administrators to better man- age the systems. To better make use of the recorded events, more practical techniques are required. From the perspective of data mining, three correlated directions are considered to be helpful for system management: (1) Provide concise yet comprehensive summaries about the running status of the systems; (2) Make the systems more intelligence and autonomous; (3) Effectively detect the abnormal behaviors of the systems. Due to the richness of the event logs, all these directions can be solved in the data-driven manner. And in this way, the robustness of the systems can be enhanced and the goal of autonomous management can be approached. This dissertation mainly focuses on the foregoing directions that leverage tem- poral mining techniques to facilitate system management. More specifically, three concrete topics will be discussed, including event, resource demand prediction, and streaming anomaly detection. Besides the theoretic contributions, the experimental evaluation will also be presented to demonstrate the effectiveness and efficacy of the corresponding solutions.

Management of remote field instrumentation via the Internet

Supervisory Control & Data Acquisition (SCADA) systems are used by many industries because of their ability to manage sensors and control external hardware. The problem with commercially available systems is that they are restricted to a local network of users that use proprietary software. There was no Internet development guide to give remote users out of the network, control and access to SCADA data and external hardware through simple user interfaces. To solve this problem a server/client paradigm was implemented to make SCADAs available via the Internet. Two methods were applied and studied: polling of a text file as a low-end technology solution and implementing a Transmission Control Protocol (TCP/IP) socket connection. Users were allowed to login to a website and control remotely a network of pumps and valves interfaced to a SCADA. This enabled them to sample the water quality of different reservoir wells. The results were based on real time performance, stability and ease of use of the remote interface and its programming. These indicated that the most feasible server to implement is the TCP/IP connection. For the user interface, Java applets and Active X controls provide the same real time access.

A Virtual Joystick as a Novel Interaction Modality for Surgeon-Computer Interaction During Computer-Assisted Surgical Procedures

Interacting with a computer system in the operating room (OR) can be a frustrating experience for a surgeon, who currently has to verbally delegate to an assistant every computer interaction task. This indirect mode of interaction is time consuming, error prone and can lead to poor usability of OR computer systems. This thesis describes the design and evaluation of a joystick-like device that allows direct surgeon control of the computer in the OR. The device was tested extensively in comparison to a mouse and delegated dictation with seven surgeons, eleven residents, and five graduate students. The device contains no electronic parts, is easy to use, is unobtrusive, has no physical connection to the computer and makes use of an existing tool in the OR. We performed a user study to determine its effectiveness in allowing a user to perform all the tasks they would be expected to perform on an OR computer system during a computer-assisted surgery. Dictation was found to be superior to the joystick in qualitative measures, but the joystick was preferred over dictation in user satisfaction responses. The mouse outperformed both joystick and dictation, but it is not a readily accepted modality in the OR.

Introduction to Security Onion

Security Onion is a Network Security Manager (NSM) platform that provides multiple Intrusion Detection Systems (IDS) including Host IDS (HIDS) and Network IDS (NIDS). Many types of data can be acquired using Security Onion for analysis. This includes data related to: Host, Network, Session, Asset, Alert and Protocols. Security Onion can be implemented as a standalone deployment with server and sensor included or with a master server and multiple sensors allowing for the system to be scaled as required. Many interfaces and tools are available for management of the system and analysis of data such as Sguil, Snorby, Squert and Enterprise Log Search and Archive (ELSA). These interfaces can be used for analysis of alerts and captured events and then can be further exported for analysis in Network Forensic Analysis Tools (NFAT) such as NetworkMiner, CapME or Xplico. The Security Onion platform also provides various methods of management such as Secure SHell (SSH) for management of server and sensors and Web client remote access. All of this with the ability to replay and analyse example malicious traffic makes the Security Onion a suitable low cost alternative for Network Security Management. In this paper, we have a feature and functionality review for the Security Onion in terms of: types of data, configuration, interface, tools and system management.

Surveillance in Cyberspace: Applying Natural and Place Manager Surveillance to System Trespassing

Research on the criminological side of system trespassing (i.e. unlawfully gaining access to a computer system) is relatively rare and has yet to examine the effect of the presence of other users on the system during the trespassing event (i.e. the time of communication between a trespasser’s system and the infiltrated system). This thesis seeks to analyze this relationship drawing on principles of Situational Crime Prevention, Routine Activities Theory, and restrictive deterrence. Data were collected from a randomized control trial of target computers deployed on the Internet network of a large U.S. university. This study examined whether the number (one or multiple) and type (administrative or non-administrative) of computer users present on a system reduced the seriousness and frequency of trespassing. Results indicated that the type of user (administrative) produced a restrictive deterrent effect and significantly reduced the frequency and duration of trespassing events.

A novel latent factor model for recommender system

Matrix factorization (MF) has evolved as one of the better practice to handle sparse data in field of recommender systems. Funk singular value decomposition (SVD) is a variant of MF that exists as state-of-the-art method that enabled winning the Netflix prize competition. The method is widely used with modifications in present day research in field of recommender systems. With the potential of data points to grow at very high velocity, it is prudent to devise newer methods that can handle such data accurately as well as efficiently than Funk-SVD in the context of recommender system. In view of the growing data points, I propose a latent factor model that caters to both accuracy and efficiency by reducing the number of latent features of either users or items making it less complex than Funk-SVD, where latent features of both users and items are equal and often larger. A comprehensive empirical evaluation of accuracy on two publicly available, amazon and ml-100 k datasets reveals the comparable accuracy and lesser complexity of proposed methods than Funk-SVD.