The author has developed an algorithm for correcting the connections between fragments of poorly formalized information received from technical systems. It is assumed that the information comes to storage and it is characterized by a finite set of indicators. These indicators are convoluted and the convolution is a weighted sum with weight coefficients. Further the resulting convolution is compared with convolutions of fragments previously received pieces of information in the storage. If the Euclidean distance between the convolutions is less than a predetermined value ε, it is believed that these pieces of information are interconnected. The connections between pieces of information are considered and corrected in the storage, i.e. the fragments of information in storage which are interconnected with up to ε, up to 2 ε, up to 3 ε etc. are identified. Thus the connection level of the first, second, third and other orders is introduced. The information that some fragments of information are connected to each other can be regarded as new information, and this new information is determined by a set of indicators, which represents a union of indicators determining the initial pieces of information. One may consider a connectivity network of the first, second, third, etc. orders, i.e. a network with elements which are interconnected up to ε, up to 2 ε, up to 3 ε, etc. In some cases it is possible to determine the connection of pieces of information, i.e. which piece of information precedes the other or which fragment follows the other. This can be done as follows. If a piece of information number I1 is defined by some indicators, as a piece of information number I2 is determined by other indicators, and if all of the first indicators are characterized by earlier time parameters than all the second indicators, the piece of information number I1 precedes the piece of information number I2. However, it is not possible to conclude that the information number I1 results from information number I2 or vice versa. To the piece of information number I2 , defined by some set of indicators, result from the piece of information number I1 , defined by a different set of indicators, it is necessary that the piece of information number I2 must already contain the piece of information number I1 , i.e. the set of indicators defining I2 must be contained in the set of indicators defining I1 . This means that the piece of information number I1 already contains the fragment of information number I2, and therefore I2 can be obtained from I1 , i.e. I2 results from I1. An example of such pieces of information is considered. Let there be a piece of information number I1 , defined by some set of indicators consisting of k elements. Then this piece of information results in pieces of information, each of which is defined by a subset of the set consisting of k indicators. The following options are possible: the pieces of information defined by various combinations consisting of (k-1) indicators of the (k-2) indicators, ..., of 3 indicators, of 2 indicators, of 1 indicator. Moreover, similarly, each piece of information defined by (k-1) indicators results in pieces of information defined by a set of (k-2) indicators, ..., 3 indicators, by 2 indicators, by 1 indicator. Similarly it is for the pieces of information defined by (k-2), (k-3) and fewer indicators. In case of a piece of information defined by a set of indicators, consisting of 3 elements, this piece of information is followed by pieces of information, each of which is determined by various combinations consisting of 2 indicators and 1 indicator. In case of a piece of information defined by a set of indicators, which consists of 2 elements, then this piece of information is followed by pieces of information, each of which is defined by 1 indicator. Thus, an algorithm was developed to set a degree of connection of various pieces of information in storage, and this connection can be regarded as new information which can be processed further.
informaciya, obrabotka informacii, tehnicheskie sistemy
1. Vorob'ev V.I., Kopyl'cov A.V., Pal'chun B.P., Yusupov R.M. Metody i modeli ocenivaniya kachestva programmnogo obespecheniya. SPb., 1992.
2. Voroyskiy F.S. Informatika. Enciklopedicheskiy slovar'-spravochnik: vvedenie v sovremennye informacionnye i telekommunikacionnye tehnologii v terminah i faktah. M., 2011.
3. Kopyl'cov A.A. Model' klassifikacii informacii i algoritm ee predvaritel'noy obrabotki dlya staticheskih i dinamicheskih ob'ektov // Izvestiya SPbGETU «LETI». Seriya «Informatika, upravlenie i komp'yuternye tehnologii». 2013. № 6.
4. Kopyl'cov A.A. Obrabotka informacii v zhivyh i tehnicheskih sistemah // Materialy HIII Sankt-Peterburgskoy mezhdunarodnoy konferencii «Regional'naya informatika - 2012» (24-26 oktyabrya 2012 g.). SPb., 2012.
5. Kopyl'cov A.A. Obrabotka slabo formalizovannoy informacii v zhivyh i tehnicheskih sistemah // Materialy Vserossiyskoy nauchno-prakticheskoy konferencii studentov, magistrov, aspirantov «Sovremennoe programmirovanie» (16-17 aprelya 2014 g.). Nizhnevartovsk, 2014.
6. Kopyl'cov A.A. Obrabotka slabo formalizovannoy informacii pri nedostatke informacii // Materialy HIII Sankt-Peterburgskoy mezhdunarodnoy konferencii «Regional'naya informatika - 2012» (24-26 oktyabrya 2012 g.). SPb., 2012.
7. Kopyl'cov A.A. Sohranenie konfidencial'nosti dannyh pri podderzhke prinyatiya resheniy na osnove izvlekaemoy special'nym obrazom informacii // Materialy VIII Sankt-Peterburgskoy mezhregional'noy konferencii «Informacionnaya bezopasnost' regionov Rossii» (23-25 oktyabrya 2013 g.). SPb., 2013.
8. Kopyl'cov A.A., Kopyl'cov A.V. Algoritm obrabotki slabo formalizovannoy informacii, postupayuschey ot tehnicheskih sistem // Izvestiya SPbGETU «LETI». Seriya «Informatika, upravlenie i komp'yuternye tehnologii». 2012. № 8.
9. Kopyl'cov A.A., Kopyl'cov A.V. Obrabotka slabo formalizovannoy informacii, postupayuschey ot tehnicheskih sistem // Vestnik Nizhnevartovskogo gosudarstvennogo gumanitarnogo universiteta. 2013. № 1.
10. Kopyl'cov A.A., Kopyl'cov A.V. Tehnicheskie sistemy i slabo formalizovannaya informaciya // Materialy HIII Sankt-Peterburgskoy mezhdunarodnoy konferencii «Regional'naya informatika - 2012» (24-26 oktyabrya 2012 g.). SPb., 2012.
11. Kopyl'cov A.A., Nechitaylenko R.A. Klasternoe atributirovanie ob'ektov informacionnoy obrabotki po ponyatiynym chastnym i integral'nym priznakam // Materialy HII Sankt-Peterburgskoy mezhdunarodnoy konferencii «Regional'naya informatika - 2010» (20-22 oktyabrya 2010 g.). SPb., 2010.
12. Kopyl'cov A.V. Ob ocenke kachestva programmnyh produktov // Problemy informatizacii (teoreticheskiy i nauchno-prakticheskiy zhurnal). 1994. Vyp. 3-4.
13. Hovanov N.V. Statisticheskie modeli teorii kvalimetricheskih shkal. L., 1986.
