2.3 Structure of information systems - IS
IP structure is a collection of its individual parts, called subsystems.
A subsystem is a part of a system that is distinguished by some attribute.
If overall structure IS is considered as a set of subsystems, regardless of the scope, then in this case the subsystems are called providing.
Among the main subsystems of IS, information, technical, mathematical, software, organizational and legal support are usually distinguished.
The structure of information systems as a set
supporting subsystems
Figure 2.3
2.3.1 Information support. Classifiers. Classification methods
The purpose of the information support subsystem is to timely generate and issue reliable information for making management decisions.
Information support is a combination of a unified system for classifying and coding information, unified documentation systems, schemes for information flows circulating in an organization, as well as a methodology for building databases.
1. Systems of classification and coding of information
classifier is a systematized set, a list of any objects that allows each of them to find their place, and have a specific (usually numerical) designation. The classification system allows you to group objects to highlight certain classes, which will be characterized by a number of common properties.
Object classification - this is a grouping procedure at a qualitative level, aimed at highlighting homogeneous properties. With regard to information, as an object of classification, the selected classes are called information objects.
In any country, state, industry, regional classifiers have been developed and are being used. For example, the following are classified: industries, equipment, professions, units of measurement, cost items, etc.
classifier - a systematized set of names and codes of classification groups.
Purpose of the classifier:
- systematization of names of encoded objects;
- unambiguous interpretation of the same objects in different tasks;
- the possibility of generalizing information on a given set of features;
- the possibility of comparing the same indicators contained in the forms of statistical reporting;
- the possibility of searching and exchanging information between various internal divisions and external information systems;
- saving computer memory when placing encoded information.
Three methods for classifying objects have been developed, which differ in different strategies for applying classification features.
Object classification methods:
- Hierarchical classification method
Taking into account a rather rigid procedure for constructing a classification structure, it is necessary to determine its purpose before starting work, i.e. what properties the objects to be combined into classes should have. These properties are further taken as classification features.
In a hierarchical classification system, each object at any level must be assigned to one class, which is characterized by a specific value of the selected classification feature. For subsequent grouping in each new class, you must specify your own classification features and their values. Thus, the choice of classification features will depend on the semantic content of the class for which grouping is required at the next level of the hierarchy.
The number of classification levels corresponding to the number of features chosen as the basis of division characterizes depth of classification.
Hierarchical classification system
Figure 2.3.1(1)
Advantages of a hierarchical classification system:
- ease of construction;
- the use of independent classification features in various branches of the hierarchical structure.
Disadvantages of a hierarchical classification system:
- a rigid structure, which leads to the complexity of making changes, since it is necessary to redistribute all classification groups;
- the impossibility of grouping objects according to previously unforeseen combinations of features.
- Faceted classification method
In contrast to the hierarchical one, it allows you to choose the signs of classification independently of each other and of the semantic content of the object being classified. Classification features are called facets(facet - frame). Each facet contains a set of homogeneous values of a given classification feature. Moreover, the values in the facet can be arranged in an arbitrary order, although their ordering is preferable.
The scheme for constructing a faceted classification system is presented in the form of a table.
|
| facets |
|||||||
| F 1 | F 2 | F 3 | … | F i | … | F n |
||
| Facet values | 1 |
|
|
|
|
|
|
|
| 2 |
|
|
|
|
|
|
|
|
| 3 |
|
|
|
|
|
|
|
|
| … |
|
|
|
|
|
|
|
|
| k |
|
|
|
|
|
|
|
|
| … |
|
|
|
|
|
|
|
|
Faceted classification system
Figure 2.3.1(2)
The names of the columns correspond to the selected classification features (facets), designated F 1 , F 2 , F 3 , …, F i , …, F n. Each cell in the table stores a specific facet value. The classification procedure consists in assigning each object the appropriate facet values. However, not all facets can be used. When building a faceted classification system, it is necessary that the values used in different facets are not repeated. The facet system can be easily modified by changing the values of any facet.
Advantages of the faceted classification system:
- the possibility of creating a large classification capacity, i.e. using a large number of classification features and their values to create groupings;
- the possibility of simple modification of the entire classification system without changing the structure of existing groupings.
Disadvantage of the faceted classification system is the complexity of its construction, since it is necessary to take into account the whole variety of classification features.
- Descriptor classification method
To organize the search for information, to maintain thesauri (dictionaries), a descriptive (descriptive) classification system is effectively used, the language of which approaches the natural language for describing information objects. It is especially widely used in the library retrieval system. The essence of the descriptor classification method is as follows:
- the population is selected keywords or phrases describing a specific subject area or a set of homogeneous objects;
- selected keywords and phrases are subject to normalization, i.e. one or more of the most commonly used is selected from a set of synonyms;
- created descriptor dictionary, i.e. a dictionary of keywords and phrases selected as a result of the normalization procedure.
Relationships are established between the descriptors, which allow expanding the area of information retrieval.
- Coding system
Used to replace the name of an object with symbol(code) in order to provide convenient and more efficient processing of information.
Coding system - a set of rules for the code designation of objects. The code is built on the basis of the alphabet, consisting of letters, numbers and other symbols. The code is characterized by: length - the number of positions in the code, and structure - the order of the symbols used to designate the classification feature in the code.
2. Unified documentation systems are created at the state, republican, branch and regional levels. The main goal is to ensure comparability of indicators of various spheres of social production. Standards have been developed where the requirements are established:
- to unified documentation systems;
- to unified forms of documents of various levels of management;
- to the composition and structure of details and indicators;
- to the procedure for the introduction, maintenance and registration of unified forms of documents.
However, despite the existence of a unified documentation system, when examining most organizations, a whole range of typical shortcomings is constantly revealed:
- extremely large volume of documents for manual processing;
- the same indicators are often duplicated in different documents;
- work with a large number of documents distracts specialists from solving immediate problems;
- there are indicators that are created but not used, etc.
Therefore, the elimination of these shortcomings is one of the tasks facing the creation of information support.
3. Schemes of information flows reflect the routes of movement of information and its volumes, the places of origin of primary information and the use of resulting information. By analyzing the structure of such schemes, it is possible to develop measures to improve the entire management system.
Example:
The simplest circuit data flows - a diagram that reflects all stages of the passage of a memo or entry in the database on hiring an employee - from the moment it is created to the issuance of an order for his admission to work.
The construction of schemes of information flows, allowing to identify the volumes of information and conduct its detailed analysis, provides:
- exclusion of duplicate and unused information;
- classification and rational presentation of information.
At the same time, the issues of the relationship between the movement of information by levels of management should be considered in detail. It is necessary to identify which indicators are necessary for making managerial decisions, and which are not. Each performer should receive only the information that is used.
4. Methodology for building databases - DB based on theoretical foundations their design. The main ideas of the methodology concept are implemented in practice in the form of two successively implemented stages in practice:
- Stage 1 - a survey of all functional divisions of the enterprise in order to:
- understand the specifics and structure of its activities;
- build a diagram of information flows;
- analyze the existing document management system;
- determine information objects and the corresponding composition of details (parameters, characteristics) that describe their properties and purpose.
- 2nd stage - construction of a conceptual information-logical data model for the field of activity surveyed at the 1st stage. In this model, all connections between objects and their details must be established and optimized. The information-logical model is the foundation on which the database will be created.
To create information support it is necessary:
- a clear understanding of the goals, objectives, functions of the entire management system of the organization;
- identification of the movement of information from the stage of occurrence to its use at various levels of management, presented for analysis in the form of schemes of information flows;
- improvement of the document management system;
- availability and use of a classification and coding system;
- possession of the methodology for creating conceptual information-logical models that reflect the relationship of information;
- creation of information arrays on machine media, which requires modern technical support.
2.3.2 Technical support of IS
The technical support of information systems is a set of technical means that ensure the operation of the IS, the relevant documentation for these tools and technological processes.
The complex of technical means includes:
- computers of any models;
- devices for collecting, accumulating, processing, transmitting and outputting information;
- data transmission devices and communication lines;
- office equipment and devices for automatic data retrieval;
- operating materials, etc.
The documentation includes the preliminary selection of technical means, the organization of their operation, the technological process of data processing, technological equipment.
Documentation can be roughly divided into three groups:
- system-wide, including state and industry standards for technical support;
- specialized, containing a set of methods for all stages of the development of technical support;
- normative-reference used when performing calculations for technical support.
2.3.3 Mathematical and software IP
Mathematical and software is a set of mathematical methods, models, algorithms and programs for the implementation of the goals and objectives of IP, as well as normal functioning complex of technical means.
The software tools include:
- management process modeling tools;
- typical control tasks;
- methods mathematical programming, mathematical statistics, queuing theory, etc.
Software tools - software include:
- General system software - these are complexes of user-oriented programs designed to solve typical problems of information processing. They serve to expand functionality computers, control and management of data processing;
- Special software - is a set of programs developed during the creation of a specific IS. It includes packages application programs, realizing the developed models of varying degrees of adequacy, reflecting the functioning of a real object;
- Technical documentation for the development of software should contain a description of the tasks, a task for algorithmization, an economic and mathematical model of the task, test cases.
2.3.4 Organizational support of IP
Organizational support is a set of methods and means that regulate the interaction of employees with technical means and among themselves in the process of developing and operating IS.
Organizational support implements the following functions:
- analysis existing system management of the organization where IS will be used, and identification of tasks to be automated;
- preparation of tasks for solving on a computer, including terms of reference for the design of IS and a feasibility study of its effectiveness;
- development of management decisions on the composition and structure of the organization, methodology for solving problems aimed at improving the efficiency of the management system.
Organizational support is created based on the results of a pre-project survey at the first stage of building a database.
2.3.5 Legal enforcement of IP
Legal support is a set of legal norms that determine the creation, legal status and functioning of IS, regulating the procedure for obtaining, transforming and using information.
The main purpose of legal support is to strengthen the rule of law.
The composition of legal support includes laws, decrees, resolutions of state authorities, orders, instructions and other regulatory documents of ministries, departments, organizations, local authorities. In legal support, one can single out a general part that regulates the functioning of any IS, and a local part that regulates the functioning of a particular IS.
Legal support of IP development stages includes regulations related to the contractual relationship between the developer and the customer and the legal regulation of deviations from the contract.
Legal support of the stages of functioning of IP includes:
- IP status;
- rights, duties and responsibilities of personnel;
- legal provisions of certain types of management process;
- the procedure for creating and using information, etc.
Definition information system(IP). Tasks and functions of IP
IP - an interconnected set of means, methods and personnel used to collect, store, process and issue information in order to achieve the goal. Computers equipped with specialized software serve as a technical base and tool for information systems.
In other words under IP is understood as an organizationally ordered set of documents (arrays of documents) and information technologies, including using the means computer science and communications that implement information processes.
Tasks and functions of IP
With organizational isolation, IS solve two groups of tasks:
1. information support task group :
selection of necessary messages and their processing,
· storage,
Search and issue of information to the subject of the main activity. (With a predetermined completeness, accuracy and efficiency in the most appropriate form for data processing systems).
2. a group of tasks related to the processing of the received information in accordance with certain algorithms or programs in order to prepare a solution to the problems facing the subject of the main activity (the so-called “ custom ” tasks).
To solve such problems IP should have the necessary information about subject area (BY) the subject of the main activity, the problems facing him, should be able to use existing models for solving problems by means of data processing or independently build such models, as well as have a certain artificial or natural intelligence.
preliminary definition. Subject area - this is a set of objects and relations between them, limited by the needs of a particular subject of the main activity.
More detailed concept « Subject area" discussed in paragraph 5 of this lecture .
To solve the tasks set, the IS should perform the following main functions:
· selection of messages from the internal and external environment necessary for the implementation of the main activity;
· information input V IP;
· storage of information in memory, its actualization and maintenance of integrity ;
· processing, searching and issuing information in accordance with the requirements set by the subject of the main activity. ( Treatment may also include preparation of solutions custom applied tasks according to the corresponding algorithms (programs)).
The composition and structure of IS, the main elements, the procedure for functioning
IP structure is an interconnected set of its parts, called supporting subsystems.
Subsystem - it is a part of the system, allocated according to some criterion.
Main supporting subsystems
· Information Support,
· technical support,
· software,
· software,
· organizational support,
· legal support.
Basic elements, the order of functioning of IS.
The main processes of information transformation are the following processes:
collection of information;
gathering information;
search and issue of information for system subscribers;
maintaining the integrity, relevance and safety of information .
These processes that ensure the work IP any purpose, can be conditionally represented as a diagram consisting of blocks:
input of information from external or internal sources;
Processing input information and presenting it in a convenient form;
output of information for presentation to consumers or transfer to another system;
feedback is information processed by the people of this organization to correct the input information.
Therefore, to implement the IS functions mentioned above, three independent functional subsystems are distinguished:
1. Organizational and technological subsystem of information collection ensures the selection and accumulation of data in the information system and includes a set of information sources, organizational and technological chains of information selection for accumulation in the system. Without a properly organized, operatively and efficiently operating organizational and technological subsystem for collecting information, it is impossible to effectively organize the functioning of the entire IP generally.
IP can process (process) only the information that is entered into it. At the same time, the quality of work IP is determined not only by its ability to find and process the necessary information in its own array and give it to the user, but also by the ability to select relevant information from the external environment.
This selection is carried out by this subsystem, which accumulates data on the information needs of users. IP(internal and external), analyzes and organizes this data, forming information profile IP. The information selection algorithm converts input streams into information array IP.
2. Subsystem for presentation and processing of information forms the core IP and is a reflection of the representation by the developers and subscribers of the system of the structure and picture of the subject area, information about which should reflect IS.
The information presentation and processing subsystem is one of the most complex components in the development IP.
This subsystem converts input information and requests, organizes their storage and processing in order to meet the information needs of the subscriber. IP.
The implementation of the functions of this subsystem assumes the presence of:
· information description apparatus , namely information - search language, coding systems and data description language;
· organizing and maintaining information (logical and physical organization, procedures for maintaining and protecting information, etc.);
· apparatus for processing and processing information (algorithms, models, etc.).
All three of these components are determined by two parameters IP: nature of information processing and functions IP.
3. Normative-functional subsystem for the preparation and issuance of information defines users, or otherwise subscribers , systems.
This subsystem directly implements the satisfaction of information needs of both internal and external users. IP. To accomplish this task, the subsystem conducts the study and analysis of information needs, determines the forms and methods of their satisfaction, the optimal composition and structure of the output information products, and organizes the process of information support and maintenance.
These functions require:
· apparatus for describing and analyzing information needs and their expressions in language IP;
· direct information support apparatus (procedures for searching and issuing information, languages for manipulating data, etc.).
If the functions performed by this subsystem are the same, IP different types they differ significantly from each other. This is especially noticeable when comparing documentary And factual IP, which will be discussed in detail later.
(internal carrier of knowledge about the subject area) is database (DB). The concept of a database is central in the field of automated information systems technologies.
Definition 1Database - a collection of data organized according to certain rules, providing general principles descriptions, storage and manipulation of data, independent of application programs.
Definition 2. (GOST): Database management system (DBMS) - a set of programs and language tools designed to manage data in a database, maintain a database and ensure its interaction with application programs.
The information core of the subsystem for the presentation and processing of information IS top level is database (BND), or automated data bank (abd) is the set of the following components :
· DB,
· DBMS,
· application components IP(a set of input and output forms, typical requests for solving information technology problems in a specific subject area),
· complex of technical means on which they are implemented.
IP classification
1. By appointment:
· reference IP,
· information support systems ,
· reference and information systems having an independent purpose.
2. By number of users and territorial basis
· single player (For example, IP, using as BY application excel ),
· low level multiplayer (For example, IP built on the application Access ),
· high level multiplayer - IP enterprise level ( distributed, very large, extra large ) .
3. By the efficiency of information processing
· real time system,
· transaction processing system,
· batch processing system .
4. By function and management levels
· production systems;
· marketing systems;
· financial and accounting systems;
· personnel systems (human resources);
5. By degree of automation
· manual,
· automatic,
· automated .
6. By the nature of the use of information
· information retrieval systems,
· information-solving systems :
o managers IP,
o advising IP.
7. By scope
· IP organizational management ,
· IP process control (TP),
· IP computer-aided design (CAD),
· Integrated (corporate) IP,
· Computing IS.
8. According to the composition of the processed information, imposing strict requirements on the apparatus for its description, organization and search
· documentaries IP(weakly structured information);
· factual IP(rigidly structured information);
· documentary and factual IP.
· geoinformation systems.
9. Scale
· world,
· international,
· republican,
· regional,
· industry,
· associations,
· enterprises and divisions.
Information Systems
3. Data models
3.2. Network model (SM)
3.3. Relational Model (PM)
4. Stages of database development
4.1. Subject area
4.2. Domain Model.
4.3. Logical data model.
4.3.1. Basic concepts
4.3.2. Relationship characteristics
4.4. Physical data model
4.5. Own database and applications
5. Designing relational databases using normalization
5.1. First Normal Form (1NF)
5.2. Second Normal Form (2NF)
5.3. Third Normal Form (3NF)
1. The concept of an information system, its structure
Information system (IS) - this is a complex consisting of an information base (information storage) and procedures that allow you to accumulate, store, correct, search, process and issue information.
Information system components:
physical component - a hardware complex on which an information system is implemented;
information component - an information database (DB) organized in a certain way;
functional component - a set of programs designed to manage the information database and documents necessary for the operation of these programs.
Note that the concept of a DBMS - a database management system - is a concept close to IS, but not identical to it. It is rather an environment and at the same time a tool for the development of information systems. The DBMS provides us with a set of procedures that facilitate the performance of typical operations on an information database.
2. Classifications of information systems
2.1. Classification by degree of automation
Depending on the degree of automation of information processes in the company management system, information systems are defined as manual, automatic, automated (Fig. 1).
Rice. 1. Classification by degree of automation
Hand ICs are characterized by the absence of modern technical means of information processing and the performance of all operations by a person. For example, about the activities of a manager in a company where there are no computers, we can say that he works with a manual IS.
Automatic ICs perform all information processing operations without human intervention.
Automated ICs involve the participation in the process of information processing of both a person and technical means, with the computer playing the main role. In the modern interpretation, the term "information system" necessarily includes the concept of an automated system.
Automated information systems, given their widespread use in the organization of management processes, have various modifications and can be classified, for example, by the nature of the use of information and by scope.
2.2. Classification based on the structure of tasks
There are three types of tasks for which information systems are created: structured (formalizable), unstructured (non-formalizable) and partially structured.
Structured (formalizable) A task is a task where all its elements and the relationships between them are known.
Unstructured (non-formalizable) task - a task in which it is impossible to select elements and establish relationships between them.
In a structured problem, it is possible to express its content in the form of a mathematical model that has an exact solution algorithm. Such tasks usually have to be solved repeatedly, and they are of a routine nature. The purpose of using an information system for solving structured problems is the complete automation of their solution, i.e., reducing the role of a person to zero.
For example, in an information system, it is necessary to implement the task of calculating payroll. This is a structured problem where the solution algorithm is fully known. The routine nature of this task is determined by the fact that the calculations of all accruals and deductions are very simple, but their volume is very large, since they must be repeated many times a month for all categories of workers.
The solution of unstructured problems due to the impossibility of creating a mathematical description and developing an algorithm is associated with great difficulties. The possibilities of using the information system here are not great. The decision in such cases is made by a person from heuristic considerations based on his experience and, possibly, indirect information from various sources.
Try, for example, to formalize relationships in your student group. You are unlikely to be able to do this. This is due to the fact that psychological and social factors are essential for this task, which are very difficult to describe algorithmically.
Note that in the practice of any organization there are relatively few fully structured or completely unstructured tasks. It can be said about the majority of problems that only a part of their elements and connections between them are known. Such tasks are called partially structured. Under these conditions, you can create an information system. The information received in it is analyzed by a person who will play a decisive role. Such information systems are automated, since a person takes part in their functioning.
For example, it is required to make a decision to eliminate the situation when the need for labor resources to complete one of the works of the complex on time exceeds their availability. Ways to solve this problem can be different, for example: allocation of additional funding to increase the number of employees; assigning the end of work to a later date, etc. As you can see, in this situation, the information system can help a person make a decision if it provides him with information about the progress of work in all the necessary parameters.
Information systems used to solve partially structured tasks are divided into two types (Fig. 2):
creating management reports and focused mainly on data processing (search, sorting, aggregation, filtering). Using the information contained in these reports, the manager makes a decision;
developing possible alternative solutions. Decision making in this case is reduced to the choice of one of the proposed alternatives.
Rice. 2. Classification according to the structure of the tasks to be solved
Information systems that develop alternative solutions can be model or expert.
Model information systems provide the user with mathematical, statistical, financial and other models, the use of which facilitates the development and evaluation of alternative solutions. The user can get the information he lacks for making a decision by establishing a dialogue with the model in the process of its study.
Expert information systems ensure the development and evaluation of possible alternatives by the user through the creation of expert systems associated with the processing of knowledge. Expert support for user-made decisions is implemented at two levels.
The work of the first level of expert support proceeds from the concept of "standard management decisions", according to which problem situations that often arise in the management process can be reduced to some homogeneous classes of management decisions, i.e. to some standard set of alternatives. To implement expert support at this level, an information fund is created for storing and analyzing typical alternatives.
If the problem situation that has arisen is not associated with the existing classes of typical alternatives, the second level of expert support for management decisions should come into play. This level generates alternatives on the basis of data available in the information fund, transformation rules and procedures for evaluating synthesized alternatives.
3. Data models
There is a wide variety of complex data types, but studies conducted on a large practical material have shown that several of the most common ones can be distinguished among them. Such generalized structures are called data models, because they reflect the user's view of real-world data.
3.1. Hierarchical model (IM)
The IM is represented by a connected graph of the tree type, the vertices of which are located at different hierarchical levels. A hierarchical database consists of an ordered set of trees; more precisely, from an ordered set of multiple instances of the same type of tree.
This model is characterized by such parameters as levels, nodes, links. The principle of operation of the model is such that several nodes of a lower level are connected by means of a connection with one node of a higher level.
A node is an information model of an element located at a given level of the hierarchy.
Let's consider IM using the example of the Our School database, which contains information about school students. From the point of view of IM, it should take the following form: the school includes classes; parallel classes are divided by letters, each class includes specific students. The model can be represented as a diagram.
The following database properties can be noted:
several lower-level nodes are connected to only one higher-level node;
a hierarchical tree has only one vertex (root), not subordinate to any other vertex;
all types of relationships must be functional (1:1, 1:M);
for the database, the complete traversal order is defined - from top to bottom, from left to right;
there is a single linear hierarchical access path to any node, starting at the root of the tree.
The most famous and widespread representative of the DBMS that implements IM is the Information Management System (IMS) from IBM. The first version appeared in 1968.
3.2. Network model (SM)
The network approach to data organization is an extension of the hierarchical one. The architecture of the network model is based on the proposals of the Conference on Data Systems Languages (CODASYL) programming language committee, 1971.
The goal of the developers is to create a model that allows you to describe M: N relationships and reduce the disadvantages of IM.
The SM database is similar to a hierarchical one; it is also based on the use of data representation in the form of a graph. From the point of view of graph theory, the SM corresponds to an arbitrary graph: in hierarchical structures, a descendant record must have exactly one parent; in a network data structure, a child can have any number of ancestors. The SM has the same basic components (node, level, connection), but the nature of their relationship is somewhat different. In SM, a free connection between elements of different levels is accepted.
As an example, consider a database that stores information about assigning subject teachers to certain classes. One teacher can teach in several classes and the same subject can be taught by different teachers.
A typical representative is the Integrated Database Management System (IDMS) from Cullinet Software, Inc.
Strengths of early (pre-relational) DBMS:
Advanced data management tools in external memory at a low level;
Ability to manually build effective application systems;
Ability to save memory by separating sub-objects (in network systems).
Flaws:
Too difficult to use;
In fact, knowledge of the physical organization is needed;
Application systems depend on this organization;
Their logic is overloaded with the details of organizing access to the database.
3.3. Relational Model (PM)
Term "relational"(from the Latin relatio - relation) indicates, first of all, that such a data storage model is built on the relationship of its constituent parts. In the simplest case, it is a two-dimensional array or a two-dimensional table, and when creating complex information models, it will be a set of interrelated tables.
The basics of the relational data model were first outlined in an article by E. Codd in 1970. This work served as an incentive for a large number of articles and books in which the relational model was further developed. The most common interpretation of the relational data model belongs to K. Date.
The relational data model consists of three parts:
Structural part.
Whole part.
manipulation part.
Structural part describes which objects are considered by the relational model. It is postulated that the only data structure used in the relational model is normalized n-ary relations.
integral part describes a special kind of constraint that must hold for any relationship in any relational database. These are entity integrity and foreign key integrity.
manipulation part describes two equivalent ways of manipulating relational data - relational algebra and relational calculus.
It is currently the most common data model supported by the vast majority of DBMSs. Typical representatives of relational systems are DB2, INGRES, ORACLE.
Consider the basic concepts of the relational data model.
The classical relational model uses only simple (atomic) data types. Simple data types have no internal structure. TO simple types data are of the following types:
Logical;
string;
Numerical.
Actually, for a relational data model, the type of data used is not important. The requirement that the data type be simple should be understood to mean that relational operations should not take into account the internal structure of the data. Of course, actions that can be performed with data as a whole must be described, for example, data of a numeric type can be added, strings can be concatenated, and so on.
In the relational data model, the concept of a domain is closely related to the concept of data type, which can be considered a refinement of the concept of "data type".
Domains– they are data types that have some meaning (semantics).
For example, domain D, meaning "employee age" can be described as the following subset of the set of natural numbers:
The difference between a domain and a subset is precisely that domain reflects semantics, defined by the subject area. There may be several domains that match as subsets but carry different meanings. For example, the domains "Part Weight" and "Quantity Available" can be equally described as a set of non-negative integers, but the meaning of these domains will be different, and these will be various domains.
The main meaning of domains is that domains limit comparisons. It is not logically correct to compare values from different domains, even if they are of the same type. This shows the semantic limitation of domains.
Attitude consists of two parts - the header of the relation and the body of the relation. The heading of a relation is analogous to the heading of a table. The relation header consists of attributes. The number of attributes is called degree of attitude. The body of a relation is analogous to the body of a table. The relation body consists of tuples. A relation tuple is analogous to a table row. The number of tuples in a relation is called relation power.
A relation has the following properties:
There are no identical tuples in a relation;
The tuples are not ordered (top to bottom);
The attributes are not ordered (left to right);
All attribute values are atomic.
Let's consider an example of the relationship "Employees" specified on the domains "Employee_number", "Last name", "Salary", "Department_number". Because since all domains are different, it is convenient to name the attributes of the relation in the same way as the corresponding domains. The relation header looks like this:
Employees (Employee_Number, Last Name, Salary, Department_Number)
Let the relation currently contain three tuples:
(1, Ivanov, 10000, 1)
(2, Petrov, 8000, 2)
(3, Sidorov, 12000, 1)
such a relation is naturally presented in the form of a table:
|
employee_number |
Surname |
Salary |
department_number |
relational database is called a set of relationships.
Relational database schema data is the set of headers of relations included in the database.
The terms that the relational data model operates on have the corresponding "table" synonyms:
|
relational term |
Corresponding "table" term |
|
Database |
Table set |
|
Database Schema |
Table header set |
|
Attitude |
|
|
Relationship header |
Table header |
|
Relationship body |
Table body |
|
relation attribute |
Table column name |
|
relation tuple |
Table row |
|
Degree (-arity) of the relationship |
Number of table columns |
|
Relationship Power |
Number of table rows |
|
Domains and data types |
Data types in table cells |
The relationship is in First Normal Form (1NF) if it contains only scalar (atomic) values.
Not first normal form can be obtained by assuming that relation attributes can be defined on complex data types - arrays, structures, or even other relations. It's easy to imagine a table with some cells containing arrays and other cells containing user-defined arrays. complex structures, and in the third cells - entire relational tables, which in turn can contain the same complex objects. Such opportunities are provided by some modern post-relational and object DBMS.
The requirement that relations must contain only data of simple types explains why relations are sometimes called flat tables. Indeed, the tables defining relations are two-dimensional. One dimension is given by a list of columns, the second dimension is given by a list of rows. A pair of coordinates (Row number, Column number) uniquely identifies a table cell and its value. If we assume that a table cell can contain data of complex types (arrays, structures, other tables), then such a table will no longer be flat. For example, if a table cell contains an array, then to access an array element, you need to know three parameter (Row number, Column number, element number in the array).
Definition 1. An information system is a set of interrelated elements that are information, human and material resources, processes that ensure the collection, processing, transformation, storage and transmission of information in organizations.
Organizations have a large number of various types of IS: from traditional to complex, based on local and global computer networks.
Definition 2. Information technology is a set of methods, procedures and tools that implement the processes of collecting, processing, transforming, storing and transmitting information.
The use of IS by firms and organizations determines the degree of modernity in the preparation of their administration to manage the organization.
Definition 3. A management information system is a range of various ISs that provide management personnel with effective decision-making on a managed object.
Definition 3 a. The management information system is a communication system for collecting, transferring, processing information about an object, supplying employees of various ranks to implement the management function
The fundamental point in determining the management information system is to ensure decision-making with its help. Management information systems are created on the basis of the study of decision-making technology using the methodology of a systematic approach. G. Simon's decision-making model can be successfully used as a conceptual basis.
According to G. Simon, the decision-making process has three stages: information, design, and also the selection stage. At the information stage, the environment is investigated, events and conditions that require decision-making are determined. At the design stage, possible areas of activity (alternatives) are developed and evaluated. At the selection stage, a certain alternative is justified and selected, organizing monitoring of its implementation. Separate stages of the process can be repeated many times if the manager is not satisfied with the information collected or the results of its processing.
At the information stage, primary data are processed and analyzed, which must be found in databases and, after appropriate processing, analyzed. Therefore, managers need to master the skills of making unplanned, situational requests, looking for the right information. The software (software) contains the corresponding powerful tools of database management systems (DBMS), as well as the necessary application packages for modeling, mathematical processing and analysis of the results.
At the design stage, the possibility of structuring a situation that requires decision-making is determined.
For structured (programmable) solutions, preliminary detailing is possible, which makes it possible to algorithmize the solution process. With the probabilistic nature of the process, the decision is determined through the probabilities of possible outcomes.
Unstructured (non-programmed) decisions arise when it is impossible to describe most of the decision-making procedures in advance. Most real situations depend on random events and unknown factors. Some procedures may be predefined, but this is not enough to automatically generate a specific recommendation. In this case, management information technologies should provide an interactive mode of operation, i.e. interactive decision support systems and expert systems that a manager can use depending on the situation.
At the selection stage, IS facilitate the choice of the right direction of activity and provide feedback to monitor the implementation of the decision. At the same time, it is assumed that at the first stages the necessary information was collected, a series of alternatives. Feedback is used to adjust the results, since the optimal solution cannot be chosen at the first step due to real time and resource constraints. To make a decision in a group mode, computer support is used, i.e. special information technologies such as IS supporting group decisions, electronic meetings, etc.
Definition 4. Decision support systems (DSS) are special interactive information management (management) systems that use equipment, software, data, a database of models and the work of managers in order to support all stages of making semi-structured and unstructured decisions directly by user managers in the process of analytical modeling based on the provided set of technologies.
Definition 5. Models are simplified abstractions of the real basic elements of the system and their relationships essential to decision making.
Information requirements directly depend on the specific level of management - strategic, tactical, operational in accordance with the functions of senior, middle and operational personnel.
Structured decisions are usually made at the operational level, tactical decisions are semi-structured, and strategic decisions are unstructured. The higher the level of management, the more unstructured decisions, so the means and methods of generating information are not the same for all levels.
At the strategic level, final ad hoc reports, forecasts and external information to develop a general strategy. At the operational level, regular internal reports are required with detailed comparisons of baseline and current indicators to help keep track of ongoing operations. Thus, information systems must meet the requirements of the respective levels and provide them with any necessary information.
Management (management) is traditionally described as a management process that includes management functions: planning, organization, personnel management, leadership (motivation) and control. IS provide the manager with data to perform all management functions.
For planning, IS provide data and models of plans ania, information about the internal state and the external environment. To support the planning function, it is necessary to have telecommunications, special problem-oriented application packages or universal modules of office systems with spreadsheets and DBMS. Software should provide “what if” analysis methods, correlation and regression analysis, statistical data processing, analysis and forecasting tools based on trends, optimization tools.
In personnel management, the most effective are information systems (IS modules) based on a DBMS, which should have an appropriate information and logical structure and allow monitoring the career and professional growth of individual employees, allowing you to process test results during periodic certification of the organization's personnel.
To manage an organization, apart from Email, there are various packages to support workflow and self-management, as well as multimedia tools for collective communication.
When exercising control without IS, it is practically impossible to develop an adequate response to a deviation from the predicted results and make adjustments to the organization's activities, therefore, when implementing an IS organization, control functions are provided in the first place.
It makes no sense for end users to constantly monitor in detail the updating and reorganization of the information technology features of the IS. Nowadays it is difficult even for data scientists. Two main aspects should be distinguished: in theory, a manager should understand so much. in order not to feel a lack of qualifications in assessing the possibilities of IP, discussing plans for their development and substantiating their opinion on this matter. In addition, the manager must be proficient in the basic methods of analysis and forecasting for the development alternative solutions, at least in spreadsheets. The role of ET in the daily work of specialists is very high. Methods of sensitivity analysis, “what if”, correlation and regression analysis, modeling and trend analysis, search for the optimal solution are implemented in spreadsheets with little or no additional programming, i.e. at the user level.
Since the real circle of end users is very diverse in business responsibilities and areas of activity, and in each case there may be special requirements, there is a universal core among all methods that can almost always help managers solve their problems.
When dealing with one object, programs running on a PC use and form a system of data about this object, usually called an information model. At first, a task-by-task approach was used, in which it was necessary to repeat the input and output of the same data. It is advisable to enter the data once and then use it in various tasks. At the same time, the independence of the process of collecting and updating (updating) data from the process of their use by software is achieved. The independence of the PP from physical organization A database achieved with the help of special (system) software that interprets a data manipulation language (procedure-oriented, not machine-oriented).
Factographic AIS, in which the databases are compiled from formalized records.
Documentary AIS, whose records may be informal documents.
Among the attributes of formatted records, there is an attribute that uniquely identifies a record. This attribute is called the primary or primary key. It determines the address of the entry in external memory.
One of the most important tasks of AIS is the rapid selection of records with certain properties. attributes. Specifying these properties. Identifies not one, but some set of records. They are called additional (secondary) keys. The search for the required records by the additional key is divided into two stages: first, the values of the main key corresponding to the records with the given value of the additional key are determined. At the second stage, according to the found values of the main key, the addresses of the records are found, and then the records themselves. For fast execution the first stage (without viewing all records in a row) use postings lists. Each list consists of pairs of values of the additional and the corresponding set of values of the main key, ordered by the additional key.
Combining the postings lists for all additional keys produces a postings file that makes it easy to find entries with the given attributes.
The main task solved in documentary AIS is the search for documents by their content. The complete solution of the search problem requires the system to understand the meaning of queries. Descriptors are some fixed set of words, including professional terms, which, in the opinion of the developer of a particular AIS, characterize the content of its documentary fund to the greatest extent. AIS looks through the text of the request in a non-formalized language and captures the descriptors encountered in the text. After that, the system looks through the full texts of all documents and selects those that contain all the descriptors found in the request. Identification of descriptors should be done up to endings. Problem: time costs. Its solution is to use the search image of the document (the list of its descriptors0. It is stored separately and has a link to the document. The search image of the query is compiled in the same way. During the search, the search images of the query and the document are compared based on the criterion of semantic correspondence fixed for the system.
A documentary AIS with simple descriptor search patterns can be considered as a factographic system with Boolean attributes equal to the total number of descriptors used. This representation is economical only for a small number of descriptors.
Organization of sequential files. The index addressing method uses a special table called an index that maps the different values of the key to the addresses of the corresponding entries. General requirements for data description languages
informational software factographic
At present, a systematic approach is used in the analysis and synthesis of large systems. The difference between this approach and the classical one, when a system is synthesized by merging components developed separately, is that a sequential transition from the general to the particular is assumed. At the heart of the system approach lies
description of the function of the system as a whole and synthesis of the elements of the system by relating the function to a specific element of the system.
Specialists in information technology deal with systems, the processes in which are associated with the processing, modification, change of information.
Information system is a purposeful set of connected elements, characterized by a certain structure and algorithm of functioning, which determines the dependence of the output characteristics of the system on the influence of the external environment and input influences.
Information systems are designed to accumulate information, store it and issue it as needed. These data are descriptions of real world objects or abstract objects that arise in various disciplines of science, and represent some true statements or messages. Over time or as a result of errors, they can become "false". Thus, one of the disciplines underlying the theory of information systems is mathematical logic.
Mathematical disciplines suitable for describing collections of objects and their properties are set theory and relational algebra (mathematical theory of relations). Information must be expressed in certain languages. For their processing on a computer, they must be expressed in formal languages (in which the meaning of sentences is uniquely determined by their form). To process information on a computer, a program must be compiled, which is a machine form of the algorithm. Finally, the processing of the program must be carried out in an acceptable time with an acceptable expenditure of systems, which is what the theory of complex systems is about.
Thus, the theory of information systems is based on mathematical logic, set theory, relational algebra, the theory of formal languages, the theory of algorithms and the theory of complex systems.
One example of information systems is a system that includes three main components:
– physical component– information system carrier ( technical means for the use of information systems);
– information component– information fund (method of organizing a system of records):
– functional component– management, updating, information retrieval, final processing.
In accordance with the defining procedures being implemented, these information systems are classified:
– information systems for technological processes. A source of information: automatic devices(sensors). Information receiver: devices, executive bodies. These systems operate in real time, that is, an unacceptably large delay in the transmission and processing of information;
– information systems of administrative and organizational type. The source and receiver of information are documents. Allowed long-term storage large amounts of information.
The classification of information systems can be represented as an administrative hierarchical, seven-level graph (Fig. 1.1).
Under structure information system refers to the totality and interaction of its individual subsystems. Each subsystem of the information system, in turn, is an information system and is characterized by a set of inputs, outputs, a law and an algorithm of functioning.
A structure often serves as a way to describe a system. At the same time, allocated systems do not necessarily correspond to physically implemented blocks or subsystems. They are chosen based on the convenience of explaining the principle of operation of the system and its features.
If further indivisible devices that form the initial elementary base are used as a subsystem, then the system will be uniquely specified using its structure. If, when describing the structure, subsystems are selected that are larger formations compared to the basic ones (i.e., indivisible further), then there is no one-to-one correspondence between the IS structure and the system itself (in the sense of its technical implementation), because one and the same algorithm, and even more so the law of functioning of such subsystems, can be implemented by means of different combinations of basic elements.
A structural description of an information system is understood as an image of the system in the form of a block diagram, i.e. collections of some blocks that have inputs, outputs and connections between blocks. In this case, the law of operation of the blocks and their input and output functions are given in a generalized form.
The difficulty of describing and choosing the structure of IS lies in the fact that in order to carry out certain operations on information, a lot of additional (service) information is created and used in the system.
The advantage of one or another IS structure essentially depends on the location of the inputs and inputs of the system.
The structure, as a method of describing a system, depends on the positions from which this description occurs, and therefore may turn out to be different for the same system.
The same system may have hierarchical structure from the point of view of information transfer and processing, and centralized from the point of view of developing control actions.
Example. Consider a system for collecting information from several sensors to a central point. Let the number of inputs, the location of sensors in space and the number of outputs be given. It is known that x i (t) is represented as continuous random functions of time, y i (t)– in the form of a sequence of numbers presented using digital indicators.
Matching rule between x i (t) And y i (t), i.e. system functioning law: y i (t)=x i (t)+Dx i ,
Where D x i- valid for i th input error. Let for each pair x i And y i the conversion algorithm is chosen, which consists in signal transmission x(t) to the center and converting it to a digital formula.
Possible options for the structure of the system: 1. Structure with independent communication lines and independent for each of the inputs of the function transformations x(t) into a numerical formula.
S- adder
TO- distribution key
3. Address system for collecting information with a common communication channel (main).
 |
W- encoder
LH- decoder
Here, information is transmitted from the i-th sensor upon request from the central point by transmitting a special address signal. To receive this signal, a decoder is located at the location of each information source. The ADC and W are also located there, allowing you to transfer the discrete value of the function over the communication line x i (t) at the time of her questioning.
