市场经济预测支持系统的智能化研究
|
摘要
近年来，随着市场经济的飞速发展和竞争的不断加剧，各个企业、
金融等部门的领导人，愈加认识到仅靠传统的经验式决策是行不通的，
正确及时的预测和由此而产生的可靠的决策，是现代企业成功的关键因
素。针对预测和决策的模型方法，已经发展到比较成熟的时期，只是并
未解决研究和应用的脱节问题；另一方面，计算机技术的不断发展，使
计算机从单纯的科学计算，发展到被应用于社会生活的各个角落，利用
计算机来实现智能化的市场经济预测，是历史发展的必然趋势。
目前已有的预测方法多散见于书刊，有关的软件系统，也多数是包
含在纯数学形式的软件包中。但企业管理者往往不是从数学模型出发而
是从企业经营的需要出发提出要解决的问题。管理者对方法、模型缺乏
全面了解，无所适从，造成应用与理论方法的脱节。本文在计算机辅助
“市场经济预测支持系统”的智能化研究方面作了如下的研究和探索：
1）研究并提出了智能化市场经济预测支持系统及其体系结构；
2）设计了系统的问题库；
3）给出了系统中模型的表示方法和管理机制；
4）给出了系统中数据库子系统的结构框架；
5）将数据仓库和数据挖掘技术应用到系统中，进一步提高系统的
智能化程度
6）对预测目标和模型的关联性进行了分析和探讨。
本文首先从体系结构、工作模型及工作流程图等方面介绍了智能化
市场经济预测支持系统的系统模型，将整个系统分为人机会话系统（人
机界面）、目标库、模型方法库、知识库、数据库、智能控制系统和分析
解释系统等部分，之后对每一部分的设计做了较详细的研究和介绍。一
般的预测系统都是对模型分类，本文的根本出发点是对预测目标进行分
类，这样就更接近用户的需求，克服脱节问题。在对预测目标分类的基
础上，适当对模型方法分类，重点研究了模型的表示法和以单元模型字
典库为基础的模型管理机制。在数据库的设计上，本文将数据仓库技术
和数据挖掘技术应用到预测系统的数据库系统当中，较好地克服了传统
数据库的缺陷，大大提高了系统的智能化程度，并给出了整个数据库系
摘 要
统的结构图。在智能控制上，本文通过五个方面来体现系统的智能化。
In recent years, with the fast development of market economy and
the
prick up of competition, leaders in the departments of
corporation and
finance are convinced that it is unfeasible to make decision oniy
by pat of
pate. The key factors for a modern corporation to success are the
right and
timely forecast and the reliable decision produced thereout.
Models and
methods aiming to forecast and decision, have run to the period
that is
relatively mature, yet the disjoint between researching and
applying hasn't
been resolved. At the same time, with the unceasing development of
computer technology, computer has passed through the period of
single
science calculate. Now computer has been applied to all corners
of society
living. Using computer to realize intellectualized forecasting of
market
economy is the unavoidable trend.
At present, the existent forecasting methods are scattered in the
books or
periodicals. The involving software is also included in the
software packages,
which are in the forms of pure mathematics. Yet managers of
corporation
always put forward the problem to be settled not from
mathematical model
but from the working need of their corporation. They are short of
full
understand of the methods and models and at loose ends. Thus
produced the
disjoint of applying and theory. The study in the aspect of
intellectualized
research of Computer Aided Market Economic Forecasting Supported
System in this paper including:
1) The system structure of Intellectualized Market Economic
Forecasting Supported System (IMEFSS);
2) The design of JMEFSS' problem base;
3) The express methods and management mechanism of models in the
IMEFSS;
4) The structure of the database sub-system;
5) Applying the technique of Data Ware and Data Mining to this
system to further improve the intellectualized degree of this
IMEFSS;
iii
ABSTRACT
6) Make some analysis on the relationship between subjects and
models.
In this paper, the system model of IMEFSS is introduced from the
aspect
of system structure, working model and working flow chart, etc.
The whole
system is divided into seven parts, which are man-machine
conversation part,
subject base, model and method base, knowledge base, database,
intellectualized control part and analysis and explain part, etc.
Then the
design of every part is detailed researched and introduced.
Sorting by the
model is the main characteristic of general forecasting system,
yet sorting by
the forecasting objects is the basic springboard of this paper,
thus it is nearer
to the user's need and it has got over the problem of disjoint.
On the base of
sorting by forecasting objects, this paper also sort on models
and methods
and mainly studied the representation of model and the manage
mechanism
of model based on unit model dictionary base. On the design of
database, this
paper firstly applied the technique of data warehouse and data
mining to it,
which got over the disfigurement of traditional database better
and improved
the system's intellectualization greatly. The structure diagram
of the system's
whole database is also given. On intellectualized control, this
paper studied
the whole system's intellectualization through five aspects.
金融等部门的领导人，愈加认识到仅靠传统的经验式决策是行不通的，
正确及时的预测和由此而产生的可靠的决策，是现代企业成功的关键因
素。针对预测和决策的模型方法，已经发展到比较成熟的时期，只是并
未解决研究和应用的脱节问题；另一方面，计算机技术的不断发展，使
计算机从单纯的科学计算，发展到被应用于社会生活的各个角落，利用
计算机来实现智能化的市场经济预测，是历史发展的必然趋势。
目前已有的预测方法多散见于书刊，有关的软件系统，也多数是包
含在纯数学形式的软件包中。但企业管理者往往不是从数学模型出发而
是从企业经营的需要出发提出要解决的问题。管理者对方法、模型缺乏
全面了解，无所适从，造成应用与理论方法的脱节。本文在计算机辅助
“市场经济预测支持系统”的智能化研究方面作了如下的研究和探索：
1）研究并提出了智能化市场经济预测支持系统及其体系结构；
2）设计了系统的问题库；
3）给出了系统中模型的表示方法和管理机制；
4）给出了系统中数据库子系统的结构框架；
5）将数据仓库和数据挖掘技术应用到系统中，进一步提高系统的
智能化程度
6）对预测目标和模型的关联性进行了分析和探讨。
本文首先从体系结构、工作模型及工作流程图等方面介绍了智能化
市场经济预测支持系统的系统模型，将整个系统分为人机会话系统（人
机界面）、目标库、模型方法库、知识库、数据库、智能控制系统和分析
解释系统等部分，之后对每一部分的设计做了较详细的研究和介绍。一
般的预测系统都是对模型分类，本文的根本出发点是对预测目标进行分
类，这样就更接近用户的需求，克服脱节问题。在对预测目标分类的基
础上，适当对模型方法分类，重点研究了模型的表示法和以单元模型字
典库为基础的模型管理机制。在数据库的设计上，本文将数据仓库技术
和数据挖掘技术应用到预测系统的数据库系统当中，较好地克服了传统
数据库的缺陷，大大提高了系统的智能化程度，并给出了整个数据库系
摘 要
统的结构图。在智能控制上，本文通过五个方面来体现系统的智能化。
In recent years, with the fast development of market economy and
the
prick up of competition, leaders in the departments of
corporation and
finance are convinced that it is unfeasible to make decision oniy
by pat of
pate. The key factors for a modern corporation to success are the
right and
timely forecast and the reliable decision produced thereout.
Models and
methods aiming to forecast and decision, have run to the period
that is
relatively mature, yet the disjoint between researching and
applying hasn't
been resolved. At the same time, with the unceasing development of
computer technology, computer has passed through the period of
single
science calculate. Now computer has been applied to all corners
of society
living. Using computer to realize intellectualized forecasting of
market
economy is the unavoidable trend.
At present, the existent forecasting methods are scattered in the
books or
periodicals. The involving software is also included in the
software packages,
which are in the forms of pure mathematics. Yet managers of
corporation
always put forward the problem to be settled not from
mathematical model
but from the working need of their corporation. They are short of
full
understand of the methods and models and at loose ends. Thus
produced the
disjoint of applying and theory. The study in the aspect of
intellectualized
research of Computer Aided Market Economic Forecasting Supported
System in this paper including:
1) The system structure of Intellectualized Market Economic
Forecasting Supported System (IMEFSS);
2) The design of JMEFSS' problem base;
3) The express methods and management mechanism of models in the
IMEFSS;
4) The structure of the database sub-system;
5) Applying the technique of Data Ware and Data Mining to this
system to further improve the intellectualized degree of this
IMEFSS;
iii
ABSTRACT
6) Make some analysis on the relationship between subjects and
models.
In this paper, the system model of IMEFSS is introduced from the
aspect
of system structure, working model and working flow chart, etc.
The whole
system is divided into seven parts, which are man-machine
conversation part,
subject base, model and method base, knowledge base, database,
intellectualized control part and analysis and explain part, etc.
Then the
design of every part is detailed researched and introduced.
Sorting by the
model is the main characteristic of general forecasting system,
yet sorting by
the forecasting objects is the basic springboard of this paper,
thus it is nearer
to the user's need and it has got over the problem of disjoint.
On the base of
sorting by forecasting objects, this paper also sort on models
and methods
and mainly studied the representation of model and the manage
mechanism
of model based on unit model dictionary base. On the design of
database, this
paper firstly applied the technique of data warehouse and data
mining to it,
which got over the disfigurement of traditional database better
and improved
the system's intellectualization greatly. The structure diagram
of the system's
whole database is also given. On intellectualized control, this
paper studied
the whole system's intellectualization through five aspects.
引文
1 赵庆祯，姜远苏等．智能化经济管理预测支持系统的研究．通过鉴定项目 2000．9
2 姜远苏，赵庆祯．市场经济预测与决策支持系统的研究．决策科学的理论、方法与应用，2000．3
3 暴奉贤编著．经济预测．经济科学出版社，1984．
4 王冠虹等．对EIS信息分析与设计的应用研究．管理信息系统，1999．6
5 李洪心．宏观经济预测分析模型系统研究．管理信息系统，1999．3
6 郭朝珍等．基于数据仓库技术和OLAP的决策支持系统．计算机系统应用，2000．3
7 游福成．轴承系列设计与分析智能专家系统．计算机系统应用，2000．3
8 刘大有等．多媒体玉米生产智能系统的研制与应用．计算机研究与发展，1999．1
9 朱顺泉．DSS的模型管理与动态构模研究．管理信息系统，1999．11
10 曹重英等，应用数据仓库技术实现决策支持系统．计算机系统应用，2000．1
11 谢榕．数据挖掘与决策支持系统．计算机系统应用，1999．8
12 何超．数据库知识发现系统及领域知识在其中的作用．计算机应用，2000．3
13 张素萍．浅论基于数据仓库的决策支持系统．计算机应用研究，1999．5
14 Internet 网上资源
15 冯文权．经济预测与决策技术．武汉大学出版社，1994．4
16 尹建生．微型计算机在统计及经济预测中的应用．地震出版社，1987．4
17 刘义亭等，现代经济与管理方法及程序．科学技术文献出版社，1989．10
18 黄锡铮．企业利润的预测．预测，1995．4
19 Fang Bi-qi, Zhou Yong, Sun Liu-quan. The Forecast and Regional Analysis of the Private Motor Vehicle Market in China. Journal of Systems Science and Systems Engineering, 1999,Vol. 8, No. 4, pp.503-513.
20 Wan Yu-cheng. Generalized Weighted Mean Combining Forecasting and Its Application in the Forecasting of Air Materials Consumption. Journal of Systems Science and Systems Engineering, 1999,Vol. 8, No. 4, pp.445-450.
21 Wang Peng-tao, Gao Xi-zhen, Zhao Li-ping. The Forecast Model of Talent Demand and Quality. Journal of Systems Science and Systems Engineering, 1999,Vol. 8,No. 3, pp.375-377.
22 樊重俊 王浣尘．企业经营计划决策支持方法与决策支持系统研究，系统工程理论与实践，1998．1
23 尤毅等．多维数据仓库设计．系统工程理论方法应用．1999，Vol．8，No．2．
24 吕亚芹．折扣组合预测法研究．预测，1999．5．
25 刘树安等．组合预测、决策模型在价格管理中的应用．预测，1997．6．
26 周传世．线性规划在确定组合预测组合权数中的应用．预测，1993．4．
27 徐大江．确定最优组合预测权重数的线性规划方法．预测，1993．2．
28 唐小我等．变权组合预测模型研究．预测，1993．3．
29 王应明．广义加权多重平均组合预测技术研究．预测，1997．1．
30 寿志勤等．三种应用环境下的预测应用系统工作模式的比较．预测，1996．5．
31 王应明等．群组预测集结方法研究．预测，1993．3．
32 W.H.Inmon.http://www.china3i.net/contribution/WDMart.htm,1996
2 姜远苏，赵庆祯．市场经济预测与决策支持系统的研究．决策科学的理论、方法与应用，2000．3
3 暴奉贤编著．经济预测．经济科学出版社，1984．
4 王冠虹等．对EIS信息分析与设计的应用研究．管理信息系统，1999．6
5 李洪心．宏观经济预测分析模型系统研究．管理信息系统，1999．3
6 郭朝珍等．基于数据仓库技术和OLAP的决策支持系统．计算机系统应用，2000．3
7 游福成．轴承系列设计与分析智能专家系统．计算机系统应用，2000．3
8 刘大有等．多媒体玉米生产智能系统的研制与应用．计算机研究与发展，1999．1
9 朱顺泉．DSS的模型管理与动态构模研究．管理信息系统，1999．11
10 曹重英等，应用数据仓库技术实现决策支持系统．计算机系统应用，2000．1
11 谢榕．数据挖掘与决策支持系统．计算机系统应用，1999．8
12 何超．数据库知识发现系统及领域知识在其中的作用．计算机应用，2000．3
13 张素萍．浅论基于数据仓库的决策支持系统．计算机应用研究，1999．5
14 Internet 网上资源
15 冯文权．经济预测与决策技术．武汉大学出版社，1994．4
16 尹建生．微型计算机在统计及经济预测中的应用．地震出版社，1987．4
17 刘义亭等，现代经济与管理方法及程序．科学技术文献出版社，1989．10
18 黄锡铮．企业利润的预测．预测，1995．4
19 Fang Bi-qi, Zhou Yong, Sun Liu-quan. The Forecast and Regional Analysis of the Private Motor Vehicle Market in China. Journal of Systems Science and Systems Engineering, 1999,Vol. 8, No. 4, pp.503-513.
20 Wan Yu-cheng. Generalized Weighted Mean Combining Forecasting and Its Application in the Forecasting of Air Materials Consumption. Journal of Systems Science and Systems Engineering, 1999,Vol. 8, No. 4, pp.445-450.
21 Wang Peng-tao, Gao Xi-zhen, Zhao Li-ping. The Forecast Model of Talent Demand and Quality. Journal of Systems Science and Systems Engineering, 1999,Vol. 8,No. 3, pp.375-377.
22 樊重俊 王浣尘．企业经营计划决策支持方法与决策支持系统研究，系统工程理论与实践，1998．1
23 尤毅等．多维数据仓库设计．系统工程理论方法应用．1999，Vol．8，No．2．
24 吕亚芹．折扣组合预测法研究．预测，1999．5．
25 刘树安等．组合预测、决策模型在价格管理中的应用．预测，1997．6．
26 周传世．线性规划在确定组合预测组合权数中的应用．预测，1993．4．
27 徐大江．确定最优组合预测权重数的线性规划方法．预测，1993．2．
28 唐小我等．变权组合预测模型研究．预测，1993．3．
29 王应明．广义加权多重平均组合预测技术研究．预测，1997．1．
30 寿志勤等．三种应用环境下的预测应用系统工作模式的比较．预测，1996．5．
31 王应明等．群组预测集结方法研究．预测，1993．3．
32 W.H.Inmon.http://www.china3i.net/contribution/WDMart.htm,1996