The effect of a database feedback system on user performance

Abstract

There are two main approaches to improving the effectiveness of database interfaces. One is to raise the level of abstraction for the content of the user-database interaction. The relational model belonging to the logical level has replaced the hierarchical and network models that belong to the lower physical level. It is likely that the relational model will eventually be replaced by models belonging to the even higher conceptual level, such as entity relationship models and object-oriented models. The second approach is to enhance the actual interaction process. This can be done by providing better feedback to the user. Feedback can be in the form of more comprehensible error messages, and the provision of a natural language interpretation of user's query. Such a feedback system was developed, and its effectiveness tested in an experiment. The results showed that the feedback system enhanced user performance greatly. Specifically, users who used the feedback system were 12.9% more accurate than those without the feedback system. They were also 41.2% more confident of their answers, and they took 29.0% less time than those without the feedback system.     

The effect of a database feedback system on user performance

There are two main approaches to improving the effectiveness of database interfaces. One is to raise the level of abstraction for the content of the user-database interaction. The relational model belonging to the logical level has replaced the hierarchical and network models that belong to the lower physical level. It is likely that the relational model will eventually be replaced by models belonging to the even higher conceptual level, such as entity relationship models and object-oriented models. The second approach is to enhance the actual interaction process. This can be done by providing better feedback to the user. Feedback can be in the form of more comprehensible error messages, and the provision of a natural language interpretation of user's query. Such a feedback system was developed, and its effectiveness tested in an experiment. The results showed that the feedback system enhanced user performance greatly. Specifically, users who used the feedback system were 12.9% more accurate than those without the feedback system. They were also 41.2% more confident of their answers, and they took 29.0% less time than those without the feedback system.     

构筑安全可靠的Web数据库应用系统

Web技术和数据库技术的结合是WWW信息服务技术发展的大势所趋 ,由此产生了一系列的安全性问题 ,本文就如何构筑安全可靠的Web数据库应用系统 ,对从工作环境到程序应用的安全性问题进行了研究。     

Building a database design laboratory on the Internet

Commercial tools for software engineering include many development environments to support the design process. These integrated tools are usually distributed through traditional retail channels, with vendors increasingly giving customers the option of downloading products from the Web. The tools are installed on user client machines and are usually duplicated for actual tool use at customer sites. Users are responsible for installation, maintenance and upgrade procedures. We developed Web Lab as part of a project sponsored by the European Commission to advance rule-based, object-oriented technologies for developing database applications. The Web Lab was designed to help disseminate the research results. It enables remote use of a design environment for database applications, including a repository for application development that provides a protected workspace. However, as software becomes more and more a commodity, we expect vendors to begin selling software use rather than tools, an option that is particularly attractive in the context of thin clients running on network computers. Web Lab demonstrates the feasibility of this idea for commercial software distribution. We describe the Web Lab project, its implementation, and plans for further development     

Building systems on relational database

Relational databases are becoming increasingly popular. They create an environment within which computer system developers can use high level tools to build application systems quickly. This article looks at one computer manufacturer, Microdata, which has built a whole product line around the use of relational database products.     

Building a terabyte NEXRAD radar database for hydrometeorology research

The management and processing of terabyte-scale radar data sets is time-consuming, costly, and an impediment to research. Researchers require rapid and transparent access to the data without being encumbered with the technical challenges of data management. In this paper, we describe a database architecture that manages over 12 TB (and growing) of Archive Level II data that is produced by the United States National Weather Service's network of WSR-88D weather radars. The contribution of this work is an automatic system for archiving and analyzing radar data that isolates geoscientists from the complexities of data storage and retrieval. Data access transparency is achieved by using a relational database to store metadata on the raw data, which enables simple SQL queries to retrieve data subsets of interest. The second component is a distributed web platform that cost-effectively distributes data across web servers for access using the ubiquitous HTTP protocol. This work demonstrates how massive data sets can be effectively queried and managed.     

Design of a distributed CODASYL database system

A general model for distributed database systems based on the three-level architecture is presented. How concrete architectural forms can be derived from the general model is illustrated by means of the CODASYL database model. A new language is introduced, the data partition language, in which all distribution specific statements are concentrated. The application of distribution predicates in the data partition language to form distribution units leads to the concept of a “distribution graph”—it is a subset of the schema graph and defines the validity scope of a distribution predicate. Global database keys, global currency indicators and multiple record access are suggested to manipulate distributed CODASYL databases.     

Hierarchical database for a multi-camera surveillance system

This paper presents a framework for event detection and video content analysis for visual surveillance applications. The system is able to coordinate the tracking of objects between multiple camera views, which may be overlapping or non-overlapping. The key novelty of our approach is that we can automatically learn a semantic scene model for a surveillance region, and have defined data models to support the storage of tracking data with different layers of abstraction into a surveillance database. The surveillance database provides a mechanism to generate video content summaries of objects detected by the system across the entire surveillance region in terms of the semantic scene model. In addition, the surveillance database supports spatio-temporal queries, which can be applied for event detection and notification applications.     

Mariposa: a wide-area distributed database system

The requirements of wide-area distributed database systems differ dramatically from those of local-area network systems. In a wide-area network (WAN) configuration, individual sites usually report to different system administrators, have different access and charging algorithms, install site-specific data type extensions, and have different constraints on servicing remote requests. Typical of the last point are production transaction environments, which are fully engaged during normal business hours, and cannot take on additional load. Finally, there may be many sites participating in a WAN distributed DBMS. In this world, a single program performing global query optimization using a cost-based optimizer will not work well. Cost-based optimization does not respond well to site-specific type extension, access constraints, charging algorithms, and time-of-day constraints. Furthermore, traditional cost-based distributed optimizers do not scale well to a large number of possible processing sites. Since traditional distributed DBMSs have all used cost-based optimizers, they are not appropriate in a WAN environment, and a new architecture is required. We have proposed and implemented an economic paradigm as the solution to these issues in a new distributed DBMS called Mariposa. In this paper, we present the architecture and implementation of Mariposa and discuss early feedback on its operating characteristics. Edited by Henry F. Korth and Amith Sheth. Received November 1994 / Revised June 1995 / Accepted September 14, 1995     

Building a database for product design knowledge retrieval—A case study in robotic design database

In product design process, when dealing with technical problems or initiating a new design, R&D personnel would often turn to technical database to seek inspiration. The building of a database with such documents has not been systematically dealt with. In this paper, several issues on how to build up a product design database are investigated: input source, sampling scheme and quality control. A case study of building a database for robotic design is used to demonstrate the concept. It is an archive of more than 1500 relevant technical papers. A total of 16 graduates are employed as operators in the labeling process and subsequently the hypothesis tests are utilized to process the labeling results. To ensure this database quality, the labeling consistency of each operator and the understanding of each category are tested. With the use of statistical methods, this work proposes a feasible and practical way to create such a database for product design.     

时滞状态反馈控制系统的稳定性增益区域

研究了一阶时滞微分方程的状态反馈P控制、PI控制问题,目的是确定反馈增益的范围使得系统的平衡态是渐近稳定的.对P控制状态反馈控制模型,利用Lambert W函数的主分支给出了确定反馈增益的显式判据以及系统的最优反馈增益;在PI状态反馈控制模型中,运用稳定性切换原理并结合D-划分法确定了在反馈增益平面上系统的稳定性区域,并利用Lambert W函数采用数值方法给出了系统的最优增益曲线.和现有方法相比较,本文方法更直观、计算更简单.     

On designing a database management system for a computer-aided engineering software system

The renaissance of engineering design can be attributed to the phenomenal growth in the microprocessor industry. The basic difference between the new and the traditional approach is one of outlook. Complex design situations involve multidisciplines. However this has made the task of integrating and transferring design information among these disciplines very difficult. Database management systems (DBMS) provide attractive solutions to these problems. A centralized control of data ensures that redundancy can be reduced, inconsistency can be avoided, data can be shared, standards can be enforced, and conflicting requirements can be handled. This paper takes the reader through the steps required to configure an engineering DBMS. The different data models are explained and their usage in CAD/CAM settings is illustrated. The importance of data independence, modularity, efficiency, and protability is emphasized.     

Prototyping Bubba, a highly parallel database system

Bubba is a highly parallel computer system for data-intensive applications. The basis of the Bubba design is a scalable shared-nothing architecture which can scale up to thousands of nodes. Data are declustered across the nodes (i.e. horizontally partitioned via hashing or range partitioning) and operations are executed at those nodes containing relevant data. In this way, parallelism can be exploited within individual transactions as well as among multiple concurrent transactions to improve throughput and response times for data-intensive applications. The current Bubba prototype runs on a commercial 40-node multicomputer and includes a parallelizing compiler, distributed transaction management, object management, and a customized version of Unix. The current prototype is described and the major design decisions that went into its construction are discussed. The lessons learned from this prototype and its predecessors are presented     

Processing real-time transactions in a replicated database system

A database system supporting a real-time application has to provide real-time information to the executing transactions. Each real-time transaction is associated with a timing constraint, typically in the form of a deadline. It is difficult to satisfy all timing constraints due to the consistency requirements of the underlying database. In scheduling the transactions it is aimed to process as many transactions as possible within their deadlines. Replicated database systems possess desirable features for real-time applications, such as a high level of data availability, and potentially improved response time for queries. On the other hand, multiple copy updates lead to a considerable overhead due to the communication required among the data sites holding the copies. In this paper, we investigate the impact of storing multiple copies of data on satisfying the timing constraints of real-time transactions. A detailed performance model of a distributed database system is employed in evaluating the effects of various workload parameters and design alternatives on the system performance. The performance is expressed in terms of the fraction of satisfied transaction deadlines. A comparison of several real-time concurrency control protocols, which are based on different approaches in involving timing constraints of transactions in scheduling, is also provided in performance experiments. Recommended by: A. ElmagarmidThis work was initiated while the author was at the Computer Science Department, University of Illinois at Urbana-Champaign.     

Principles of database management in a distributed system

The principles of database design in a distributed system are examined. The organization of the data, derived from a close analysis of data requirements and data usage, is the most important design decision. The type of system depends on the type of database and so does the manner in which decentralization is achieved.While only loose data integration is currently feasible in a distributed system, central control and standar dization is a valuable investment for the future.     

HGDBMS: a human genetics database management system

Human genetics research involves a large number of complex data sets naturally organized in hierarchical structures. Data collection is performed on different levels, e.g., the project level, pedigree level, individual level, and sample level. Different aspects of a study utilize different views of the data, requiring a flexible database management system (DBMS) which satisfies these different needs for data collection and retrieval. We describe HGDBMS, a comprehensive relational DBMS, implemented as an application of the GENISYS I DBMS, which allows embedding the hierarchical structure of pedigrees in a relational structure. The system's file structure is described in detail. Currently our Melanoma and Chromosome 17 map studies are managed with HGDBMS. Our initial experience demonstrates the value of a flexible system which supports the needs for data entry, update, storage, reporting, and analysis required during different phases of genetic research. Further developments will focus on the integration of HGDBMS with a human genetics expert system shell and analysis programs.     

Approximate pattern matching in a pattern database system

This work is concerned with the organization of a large database of binary pictures (normalized for size, rotation and position) and with the efficient “inexact match” of an input pattern against the whole database. Current techniques in pattern analysis use matching algorithms without any regard to the global organization of the storage representations of the models they are to match. Consequently, the algorithms are only practical for small databases. This paper discusses the design of a pattern database system and the economy that it provides for the matching problem. The database organization is based on the quad-tree representation of binary patterns. Given an arbitrary decomposition D (or partition) of a pattern P and an arbitrary function f on the pattern, we repeatedly apply f on D(P), D(D(P))P, … to obtain higher and higher levels of abstraction f(D)(P)), f(D(D(P))), … of the pattern. The computed values obtained after the jth application of f are used to label the ith level of the pyramid. The specific representation used in this paper is called the sum-quad-tree, in which each level of the tree stores the sum of the labels of its sons. The lowest level of the sum-quad-tree corresponds to the individual pixels and is the nth level (i.e. node m at level n implies v(m) = 0 or v(m) = 1). Nodes at the jth level of the sum-quad-tree correspond to sums of 2^n?j × 2^n?j picture points, so that the 0th level contains the number of 1's in the pattern. The pyramid representation is used as a hierarchical (n-level) indexing scheme of the database. The advantage of this organization is that the matching algorithms presented reject most of the patterns in the database by utilizing the relatively small in size index tables and thus avoid the overhead of unnecessary CPU time and $\text{I}\text{O}$ operation between main memory and secondary storage.     

Boundary feedback stabilization of a rotating body-beam system

This paper deals with boundary feedback stabilization of a flexible beam clamped to a rigid body and free at the other end. The system is governed by the beam equation nonlinearly coupled with the dynamical equation of the rigid body. The authors propose a stabilizing boundary feedback law which suppresses the beam vibrations so that the whole structure rotates about a fixed axis with any given small constant angular velocity. The stabilizing feedback law is composed of control torque applied on the rigid body and either boundary control moment or boundary control force (or both of them) at the free end of the beam. It is shown that in any case the beam vibrations are forced to decay exponentially to zero     

Stabilizing a linear system with quantized state feedback

The problem of stabilizing an unstable, time-invariant, discrete-time, linear system by means of state feedback when the measurements of the state are quantized is addressed. It is found that there is no control strategy that stabilizes the system in the traditional sense of making all closed-loop trajectories asymptotic to zero. If the system is not excessively unstable, feedback strategies that bring closed-loop trajectories arbitrarily close to zero for a long time can be implemented. It is also found that when the ordinary linear feedback of quantized state measurements is applied, the resulting closed-loop system behaves chaotically. The asymptotic pseudorandom closed-loop system dynamics differ substantially from what would be predicted by a conventional signal-with-noise analysis of the quantization's effects. Probabilistic reformulations of the stability problem in terms of the invariant measure are considered