In the left part of a window there is a current structure of the project in which you can allocate those articles and categories which wish that participated in structurization process.
In right — the future structure of the project which to you at first is necessary for describing. For this purpose using the contextual menu categories and incorporated records are created. What is the categories it is not necessary to explain, I will explain just myself represent records. incorporated records are records in which on the keywords chosen by you texts of suitable single records will be located, i.e. result will be creations of one big record from group small.
Distribution can be uniform or релеватным. During uniform distribution the program tries to scatter more or less in regular intervals articles on those elements of structure in which it approaches. If it is used релеватное distribution — the program counts up quantity of coincidence of each of the specified keys and sends article in that element of structure, in for which the total quantity of keys has given most of all coincidence.
Under a window with the structure description there are fields of management of separate record of this structure:
- the Name — the name of the future category or incorporated record
- Keys — templates of keywords on which articles will be certain in a concrete category.
- Fields for search — where exactly the program should search for keys
Operating symbols in keywords:
- If the template begins with ^ — then it means that the text should begin with words following this symbol (conveniently for creation of indexes, for example: a screenshot ).
- If the template begins on + is means that the word necessarily should be present at the text
- If the template begins on — is means that the word should not be in the text
the Cap of article for incorporated record is a divider of articles which will be present at the beginning of each separate record. For example, if at you a resultant record consists of 3 which records of the name «Record 1», «Record 2» and «Record 3», and cap options same as are shown on a screenshot the initial code of record will look as follows:
<h3> Record 1 </3> ... The text of record 1... <h3> Record 2 </3> ... The text of record 2... <h3> Record 3 </3> ... The text of record 3...
After creation of structure and filling of all fields it is necessary to press the button to Analyze — which will spend estimation and will show in what category how many articles will get after application.
a distribution Principle
That it was easier to understand as well as where your articles after structurization will get it is necessary to describe a principle on which program works:
- At first the allocated records in the left window are distributed between the first level of the future structure. If one record approaches for several categories it gets to that category, in which least records. It allows to reach uniformity of distribution
- Further record of each category of the first level are distributed between affiliated elements (i.e. those records participate in distribution only that on the previous step have got in parental). If there are records which do not approach for one affiliated element (by an element mean a category или объединенную запись) these records remain in this category.
- distribution of records between elements of the bottom levels by a principle described above is Following step.
the Insert of categories/records from the buffer
Here all is very simple, records in the buffer should be in such format:
The name; Keys; Fields for search The name; Keys; Fields for search ... The name; Keys; Fields for search
And the last 2 are not obligatory. If to insert so:
The name; Keys The name; Keys ... The name; Keys
That weeding for search will be for all these records «Under the name стати». And if so:
the Name The name ... The name
Then also keys will be empty.
Divider of keywords (if them for each record should be a little) — a comma.
Structurization allows the user to define some criteria of a choice of records for reception of more exact results. For each of types of records (categories and incorporated records) are the options maximum and the minimum quantity of records. If at use of the user options the program вылазит for these borders — then it starts to manipulate options «the Field for search» to achieve more comprehensible results.
it is necessary to know that in view of that full search of every possible variants very much ресурсоемкий and long process therefore the algorithm of search of an optimum variant is developed so that not to take away a lot of time and thus to have comprehensible efficiency.
If the program for record an exit for borders minimum or maximum значние then it changes options «the Field for search» by a following principle:
- If value is less, than it is necessary — that «the Field for search» changes on previous (that that above) from the list specified by the user «the Priority of fields for search»
- If value more than is necessary — that «the Field for search» changes on the following (that that more low) from the list specified by the user «the Priority of fields for search»
It follows from this that "for search " it is necessary to specify the Priority of fields so that that field which will bring more records were above, and those that less — more low.