DE10017608A1 - Record processing method for a computer data-bank system, involves initially arranging the objects from their data records into a comparable description of selectable features - Google Patents

Abstract

Processing of operations in a data bank system holding an arbitrary number of data records in a computer store involves arranging the objects from their data records into a comparable description of selectable features (M1,M2...,Mn), followed by storing the specific record sets (M1,1, M1,2....M1,n, M2,1, M2,2...M2,n to Mn,1,Mn,2....Mn,n) in a features zone/area (MB). The relationships between the detected objects and the respective record sets of the selectable features are produced by the data bank internal index.

Description

The invention relates to a method for performing operations in a database system where one consisting of a large number of data records Database is stored in a data structure with access to this data and its display.

It is well known and general in data processing usual, objects of different types, such as products, Services, procedures, etc., by assigning their Data (characteristics, properties, technical and con structural parameters, etc.) for a clear identification term fen (counter, manufacturer, type numbers, etc.) and in a machine-readable memory to save. The sequence of these data forms the description sentence of each object. To  this data is common for their primary use clear within the description sentence, in particular divided into groups. The data of comparable Ob objects are usually grouped together in object groups everyone and also in public databases offered. For the search (selection) only a few the selection-determining, a comparison possible corresponding data of the description sentence about special Query functions of the database systems used. The Realization of such databases requires one relatively large expenditure of specialists. The permissible Query functions are usually by the user not freely selectable, but are made by the database operator driver specified. Even with the same object groups some of them are large between the operators Differences on.

When querying relational databases with the Query language SQL can be multiple data (criteria en) link logically to selection conditions, at for example, by AND or OR conditions. Around to be able to evaluate such and other conditions are generally derived from the original tables per query date again tables as intermediate results won who, according to the operators of the interrogation practice be linked to the results table.

Methods of iterative query are also known, where a first result table with further out polling conditions is queried, so z. More colorful www.weltauto.com. The necessary high effort for that Sorting the lists and an insufficient ant word-time behavior of these database technologies be limit user acceptance. The data for the tables are often in unsorted form, if, for example, one sor after a secondary index tated table the ident terms of the concerned  Contains records as data and in a selection condition is included according to value ranges. At property group pen with a large number of selection-defining characteristics le are search masks and filter techniques in the Practice therefore on some of these features borders. This leads to unsatisfactory results Results since the query is either in the form "no result" or "multiple results" ends.

The prior art shown at Database technologies is widely used of data processing is unsuitable, because the for operational applications of databases typical Methods of searching for data on one Object only to search for several ver similar, in the characteristics of the search mask extended objects, but that for object group-typical comparison of the data of objects is not supported.

The object of the invention is therefore to provide a method for Execution of operations in a database system Provide one of a variety of Existing database in a data structure tur is stored and which if the presented disadvantages a simple and flexible Access from the stored data in any order on the objects and their display ensures.

According to the invention the object is achieved by a method, the essence of which is that an arbitrary number of data records of comparable objects is stored in a computer and the operations are carried out on the basis of a data structure in which the objects from their data records are used to determine their comparative description Characteristics (M ₁ , M ₂ ... M _n ) are assigned in a description sentence, the specific expressions (M _1.1 , M _1.2 .. M _{1, n} , M _2.1 , M _2.2 M _2, n to M _{n, 1} , M _{n, 2} ... M _{n, n} ) each form the selection-defining data record of the objects and all of them object-dependent characteristics (M _1.1 , M _1.2 ... M _{1, n} , M _2.1 , M _{2.2 ...} M _{2, n} to M _{n, 1} , M _{n, 2} ... M _{n, n} ) of the characteristics (M ₁ , M ₂ ... M _n ) are each combined and stored in corresponding feature areas (MB), the relationships between the detected objects and the respective characteristics of their selection-determining features ch internal database pointers are mutually produced and these are saved and finally the display of stored data by direct access to the number of the respective object or the direct and step-by-step selection of the characteristics (M _1.1 , M _1.2 . , , M _{1, n} , M _2.1 , M _2.2 . , , M _{2, n} to M _{n, 1} , M _{n, 2} . , , M _{n, n} ) of the selection-defining features (M ₁ , M ₂ ... M _n ) takes place.

Advantageous embodiments of the invention Procedures result from the features of the claims che 2 to 8.

The method according to the invention organizes selections matching data on objects within object group pen such that different forms too Characteristic areas summarized in tables (lists) become. The object data, the within an object group for identification of each object in this group sufficient and for a comparison (the presentation of the deviations / Un differences) of each object with each other necessary are dig. That is, is every object in the object group with at least one expression in each characteristic area richly represented and different from all  other objects in at least one form of its characteristics. The number of occurrences is noticeable depending on the time and any.

The sequence of the names of the characteristics thus forms the object description record of an object group; the stringing together of an expression Each table gives the description of an object the object group.

The invention also arranges by means of a pointer any reference to an exclusion within defi a corresponding rever assign. For object groups organized in this way, this means pen that on each sentence number (ident term) every applicable expression and of these characteristics refer back to all applicable sentence numbers will.

In the following the invention with the aid of figures are explained in more detail. Show it

Fig. 1 is a graphical representation of the relationships between objects and characteristics in areas of characteristics;

Fig. 2 is a graphical representation of the relationships between objects and characteristics in the characteristic areas and by means of an intermediate area for the characteristics of the features areas;

. Fig. 3 is a graphical representation according to FIGS 1 and 2, supplemented by the relationships to the parent in a feature portion detected occurrences;

FIG. 4 shows a graphical representation according to FIG. 1 with an active and inactive state form of a feature area.

Fig. 1 illustrates by means of database internal pointer, the relationships between the objects 1, 2, 3, and n and the characteristics of the feature portion MB ₁ (M _1,1 to M _{1, n),} the feature portion MB ₂ (M _2,1 to M _{2 , n} ) and those of the MB ₃ feature range (M _3.1 to M _{3, n} ).

Thus, the pointers 2 1, 2, ₂ , ₃ and _{3 3.2 shown} for the object express that the object 1 with the characteristic M _1.2, the characteristic M ₁ , with the characteristic M _2.3, the characteristic M ₂ and with the characteristic M _{3.2 is} assigned to the characteristic M ₃ , as vice versa in the characteristic area MB _1, the characteristic M _1.2 applies to both the object 1 and the object 2 and the characteristic M _{1.3 is} the object 3 concerns.

From Fig. 2 it can be seen that in addition to the tenbank internal pointers according to Fig. 1 for special applications of all forms in all features areas pointers to all forms of objects can be stored in the other feature areas. This enables selection processes without access to the list of objects, for example in such a way that from the characteristic M _1.1 in the characteristic area MB ₁ to the characteristics M _2.2 in the characteristic area MB ₂ and M _3.3 in the characteristic area MB ₃ Reference is made, which is supplemented by the expressions M _4.1 ; M _4.3 and M _4.4 in the MB ₄ feature area form the data record for object 4. The characteristic area MB _{4 is} a summary of "alternative" forms M _4.1 to M _{4.5 of} the objects and part of a multi-level branching in the memory structure, ie the characteristics in the characteristic area MB ₄ are accessed in both directions via the database NEN intermediate area ZM, in the occurring combinations of features of the MB ₄ feature area as features ZM _4.1 ; CM _4.2 ; CM _4.3 . , , ZM _{4, n} are stored.

For reasons of clarity, FIG. 2 does not show graphically that the characteristics of the feature areas MB ₁ ; MB ₂ ; MB ₃ and MB ₄ are linked together. Only the corresponding pointers were assigned to the values.

The creation of an intermediate area ZM or more Intermediate areas make sense, for example, if for a large number of objects only a few Combinations of "alternative" forms is applicable and therefore shorter access times are made possible and storage capacity is saved.

FIG. 3 shows that, in addition to the data structure shown in FIG. 2, individual or more expressions from one or more characteristic areas are grouped together under superordinate aspects into groups which here carry the identifier "O", the superordinate characteristics O _{1, 1} -O _{1, n} as characteristics in the higher-level feature area MBO ₁ are shown. This means, for example, that the higher level O _1.4, the levels M _2.1 ; M _2.2 -M _{2, n} summarized and referred to by means of the corresponding pointer.

FIG. 4 shows a data structure analogous to FIG. 1, in which the feature area MB _{3 has} an active and an inactive state form. The pointers of the active form Z _a3.1 ; Z _a3.2 ; Z _a3.4 . , , Z _{a3, z} indicate the relevant expressions M _3.1 ; M _3.2 . , , M _3.1 in the MB _{3 characteristics area} , to which the corresponding references to the objects are assigned. The state forms can be changed by external criteria and so the pointers Z _i3,2 ; Z _i3.1 ; Z _i3.3 . , , Z _i3.4 activated and the pointers of the previously active form are deactivated.

According to the inventive method are under associated data to understand all object data that are not determining selection. That is in particular those that only target one object within the group apply (e.g. property number, manufacturer name, Photo etc.). They continue to be in well-known tables len- or database systems saved.

The description of objects generally results from a sequence of selection-defining characteristics (properties), the different forms (values) of which are usefully combined into characteristic areas (tables). Objects with the same sequence of features can therefore be arranged and saved in an object group in such a way that

• - an object-dependent number of characteristic areas with at least all within the object group appearing occurrences,
• - for all contained objects that apply in each case the values determined and in a sequence of Pointers (references) to these characteristics Object number saved and
• - each version the pointers (references) to all the contained objects are assigned in their Description contain the respective expression is.

Deleting or adding objects always means a change in the sequence of pointers, u. May also an extension of the values in the characteristics pass. This update can be done through a program from existing tables or relational databases ken or by simply clicking all for each due to the relevant object.

The method according to the invention has the special one Preferred that with the data structure described direct access without sorting with either Object number on all applicable data (type gen) as well as with an appropriate sequence of data Is possible on an object number. The However, the procedure also leads directly to any (valid) expression for all in other characteristic ranges valid (permissible) off embossing and thus enables the user to calculate managed, gradual and very flexible restrictions selection solutions. It stays with everyone Selection step the representation of all (still) in the Obtain characteristics contained in the object group, d. H. the sum of the selection decisions made by the user and the result of these decisions the restrictions in all characteristic areas - if no variants are available for selection are - always the result of expressions to one Object.

For certain applications, an extension of the Data structure can be meaningful for all expressions next to the pointers to all applicable objects also the pointers to all applicable versions of these objects in the other characteristic areas be saved. In such cases, e.g. Legs Selection of any objects within the characteristic area rich possible on a computer (client) and only  after the selection of the access to the example have a database ordered by object numbers another computer (server) to supplement the Ob Project data necessary for the assigned data.

The different forms of object descriptions Exercises can have an alternative character (black or white or. . .) or without alternatives (available or not available). That is why for object groups in which a multitude of such "Alternative" forms appear, their together Version of feature areas with now alternative Combinations of these forms and a multi-level Ge storage in the storage structure. While So in all feature areas with alternative off Embossing only one specification for an object description exercise applies, apply to the "internal" memory combinations and all objects (Features) of such alternative forms with the selection of an expression in the corresponding Characteristics area already several other forms as "elected".

A special form of multi-stage decay represents the grouping of forms and the Assignment to "generic terms", for example for a rough and subsequent fine selection. Typical Examples can include the grouping of car models the manufacturers or from color gradations to the Be primary colors. For the storage structure means this is that with one or more expressions conditions in a characteristic area or several characteristics all areas of a "subordinate ten "feature area. The corresponding The corresponding expression must be subordinate for all objects th range of features apply, but does not have to Be part of the description set of the objects. The  is especially in such applications expedient, in which real forms in the form of Numerical values over several levels according to different Algorithms summarized into value ranges and for a multi-stage pre-selection should be offered.

Individuals can be used in the description set of objects Features one active and one or more inactive states with the same or under have different forms. It applies to them that for each condition in the respective feature area the corresponding pointers are saved for the values be saved and that the decision to activate one of these states for the object description Exercise in general on criteria regardless of Be notation, but during the selection process is to be met by the user. One of the state forms is determined at the beginning of a selection. Typical example le for this can be quantity or season dependent Prices be.

It should be noted that in the memory structure ren for object groups the described forms of Characteristic areas in the "pure" form (all characteristic areas have the same shape), but also in the "Mixed form" (any number of feature areas from any of the forms) can occur. Likewise, remains to note that to designate an object group, to designate all or some of their characteristics rich and / or to all or individual forms associated data can be saved if this data is the same for all applicable objects are, in particular also data that the selection characterize user behavior.  

In the memory structure can be any feature areas are included or retained to those who are first or from a certain No pointers are assigned at the time. they should in the layout of the relevant characteristic areas corresponding to "documented" characteristics and make z. B. recognizable that objects with these Aus embossments were already or expected in the offer become.

The process according to the invention is intended to be described below Based on two examples, a rental offer apartments, to be explained in more detail.

Fig. Shows that the data set consists of an object from 5 following features:
M ₁ - number of the apartment;
M ₂ - city district;
M ₃ - number of rooms;
M ₄ - applicable floor;
M ₅ - size of the apartment;
M ₆ - rental price;
M ₇ - elevator equipment;
M ₈ - balcony equipment;
M ₉ - parquet equipment;
M ₁₀ - part no. Landlord;
M ₁₁ - Name of landlord.

The characteristics of these features are shown in the feature areas MB ₁ to MB ₁₁ .

According to the invention, the characteristics M ₂ to M _{9 are} regarded as selection-determining from FIG. 5 by means of pointers internal to the bank, the characteristics M ₂ to M _{6 representing} “alternative” forms and the characteristics M ₇ to M ₉ in FIG. _{6 a} are summarized as "alternative" expressions in the intermediate area ZM.

Apartment No. 1 is determined as follows:
Characteristic M ₂ - through the expression M _2.2 , Köpenick;
Feature M ₃ - through the expression M _3.3 , 5 rooms;
Feature M ₄ - through the expression M _4.3 , 3rd floor;
Feature M ₅ - through the expression M _5.4 , 95 m ² ;
Characteristic M ₆ - through the form M _6.3 , 1,400.00 DM;
Feature M ₇ - due to the M _{7.2 design} , elevator;
Characteristic M ₈ - through the expression M _7.1 , balcony;
Characteristic M ₉ - through the expression M _7.3 , parquet.

Alternatively, the features M ₇ to M _{9 can be represented} as a combination of K _1,2,3 , K _1,2 , K _1,3 and K _2,3 in the intermediate area ZM.

Thus, for the characteristic M ₂ - city district - in the characteristic area MB _2, the characteristic M _2.1 of objects 3, 7 and 12, the characteristic M _2.2 of objects 1, 5 and 10, the characteristic M _{2.3 are} Objects 8 and 9 and the expression M _{2.4 of} the objects 2, 4, 6 and 11 Darge represents.

The number of the apartment and the names of the landlords and their phone numbers are considered non-selection rated for the apartment seeker. she can select the result after selecting an apartment be classified.

In the event that the cheapest apartment is sought, the rental price of DM 1,320.00 would have to be chosen. As can be seen from FIG. 7a, this selection criterion applies to apartments in the city districts of Köpenick and Lichtenberg, for an apartment with 4 rooms and an apartment with 4½ rooms, apartments on the 2nd and 4th floors, apartments with a size of 91 m ² and 97 m ² as well as with balcony and / or elevator and / or parquet too. The relevant values are shown here without hatching. The selection of every further variant would lead to a result without having to go through any further selection steps.

Fig. 7b shows the housing seekers the values of the non-selection determining characteristics.

In another example, which is shown in FIGS. 8a, 8b, 8c and 8d, a 5-room apartment is to be sought from the objects according to FIG. 5.

As can be seen from Fig. 8a, there is such an apartment in the districts of Friedrichshain, Köpenick and Treptow, each on the 1st to 3rd and 5th floors, in an apartment size of 93, 95, 102 or 108 m ² , with a rental price of 1,380.00 DM, 1,400.00 DM or 1,450.00 DM and with balcony and / or elevator and / or parquet. FIG. 8b shows the Ausprä conditions of the respective non-selection determining characteristics.

According to Fig. 8c, when choosing an average rental price of DM 1,400.00 as the next criterion, there are still freely selectable types of apartments in the city districts of Köpenick and Treptow, apartments on the 3rd or 5th floor, apartments with 93 m ² or 95 m ² as well as with elevator and parquet and (still selectable) balcony. The selection of a further variant would then lead to a single result. From Fig. 8d the values of the corresponding non-selection determining characteristics in turn are illustrated.

The examples show that the inventive method enables a person looking for a flat to very simple and flexible way according to his Needs and considering his poss options for choosing a home that suits him approach.  

The universality of the method according to the invention allows its applicability for the difference most diverse industries, for example for the hotel and Catering as well as for the real estate industry and the car trade.

Claims (8)

1. A method for performing operations in a database system in which an arbitrary number of data records of comparable objects are stored in a memory of a computer, characterized in that the operations take place on the basis of a data structure in which

• - The objects from their data sets for their comparative description are assigned characteristics determining characteristics (M ₁ , M ₂ ... M _n ) in a description set, the specific characteristics of which (M _1.1 , M _1.2 ... M _{1, n} , M _2.1 , M _2.2 ... M _{2, n} to M _{n, 1} , M _{n, 2} ... M _{n, n} ) form the selection-determining data record of the objects;
• - the specific object-dependent characteristics (M _1.1 , M _1.2 .. M _{1, n} , M _2.1 , M _2.2.. M _{2, n} and M _{n, 1} , M _{n, 2} . M _{n, n} ) of the features (M ₁ , M ₂ ... M _n ) are each combined and stored in a feature area (MB);
• - The relationships between the detected objects and the respective characteristics of the selection-determining features are mutually established by internal pointers and these are stored;
• - The display of stored data by direct access to the number of the respective object or the direct and step-by-step selection of the characteristics (M _1.1 , M _1.2.. M _{1, n} , M _2.1 , M _2.2 M _{2, n} to M _{n, 1} , M _{n, 2} ... M _{n, n} ) of the features determining the selection (M ₁ , M ₂ ... M _n );
• - Non-selection-defining features of the objects are stored in conventional database or table systems.

2. The method according to claim 1, characterized in that that the selection-defining data record from the characteristic areas (MB) with alternatives (exclude each other den) and / or alternative (not out closing) forms is formed. 3. The method according to claim 1 or 2, characterized in that a sufficient number for the identification of an object selection-determining features (M ₁ , M ₂ ... M _n ) is determined. 4. The method according to any one of claims 1 to 3, characterized in that each object in each feature area (MB ₁ , MB ₂ , MB ₃ to MB _n ) with at least one expression (from M _1.1 , M _1.2 ... M _{1, n} ; M _2.1 , M _{2.2 ...} M _{2, n} or M _{n, 1} , M _{n, 2} ... M _{n, n} ) is represented. 5. The method according to any one of claims 1 to 4, there characterized by that internal database pointer each from an expression from the feature description rich (MB) of an object on the characteristics in refer to the other areas of characteristics, that apply to the object and vice versa. 6. The method according to any one of claims 1 to 5, there characterized in that no alternative conditions in the data set of objects in an intermediate feature area (ZM) are arranged in which they in combination form new forms that can be assigned to objects using pointers. 7. The method according to any one of claims 1 to 6, characterized in that expressions of a feature area or several feature areas in a higher-level feature area (MBO) or in several higher-level feature areas (MBO ₁ , MBO ₂ to MBO _n ) are grouped together as summary expressions , where internal database pointers from the individual versions of the data record refer to the relevant summary in the higher-level area (MBO) and, on the other hand, pointers to the objects, the versions of which are recorded in the respective higher-level area and the links are stored. 8. The method according to any one of claims 1 to 7, there characterized in that the record of a Object for one or more characteristics contains several forms, each of which only an expression for the person who determines the selection Record is activated, using the pointer on the individual active and inactive characteristics gene and vice versa.