Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1998-10-29
2001-10-02
Black, Thomas (Department: 2171)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000, C707S793000
Reexamination Certificate
active
06298350
ABSTRACT:
FIELD OF THE INVENTIONS
The suggested inventions relate to information technologies exactly to processing information materials to prepare them for personified use that is to the processing oriented on a particular user or a group of users having definite, supposed as known, needs in information.
THE BACKGROUND ART
There are wide known traditional methods for processing information which is contained in materials' flow to prepare thereof for using by definite customers including methods for producing abstracts and summaries as secondary materials which characterize or present shortly contents of primary materials.
Such methods not suggesting any formalized principals for evaluating the content of primary materials have drawbacks due to human factor in individual approaches of specialists affecting on the processing and correspondingly on results.
There are also known methods for processing information in materials' flow by listing preselected hint words and producing an abstract of the document by juxtaposing extracted sentences containing such words. (See as an example European Patent application 0 361464 A2 with date of publication 04.04.90.)
In similar methods there is more probable, than in the methods of the previous group, the selection of only that information which is connected with the information needs of the user defined by a list of reserved words. However, missings of wanted information and false hits are possible since words can have more than one meaning and the same meaning can be expressed by different wordings.
Moreover, a number of methods is known which are grouped by directing on detecting a “latent structure” of information being interesting for the user on the base of an analysis of his previous demands. (See as an example Scott Deerwester et al., “Indexing by Latent Semantic Analysis”, Journal of the American Society for Information Science, 41(6), p. 391-407, September 1990.)
In these methods which are also oriented on detecting coincidences of search images with elements of the processed materials' content due to more complicated composing of the indicated images forming so called “user's profile”; i.e. his information needs' characteristic, errors, inherent in methods of the previous group, are less probable.
However, in this and similar methods, the stress is made on forming “user's profile” making this in advance and not paying attention to obtaining an integrated presentation of the processed materials' content as well as to the content of one multiaspect document of great volume during processing itself since such methods including methods of the previous group are oriented on an independent analysis of separate materials of the processed flow. Moreover, although the noted stress creates some background for automation of such methods it is not favorable for full activating abilities of specialists dealing with “hand” processing the flow of information materials.
THE DISCLOSURE OF INVENTION
The claimed inventions are presented by six methods-embodiments for processing information materials for personified use.
The most close to all claimed methods is the last method above described by S. Deerwester et al.
Each of the claimed methods solves the task to obtain a technical result concluded in activating psychophysical means affecting (stimulating) associative (interpretive) abilities of a specialist dealing with processing information materials' flow, on an appropriate consideration of user's information needs to create conditions for personified information support. The technical result achieved using the present methods is concluded also in providing to a user the content of the whole complex of the processed information materials in the integrated form with a quantitative evaluation of relative covering, in processed materials, various components of “user's profile” both separately and in relation with a conclusion of sequenced quantitative characteristics of a suggested integrated form that permits to determine a sequence of acquaintance with information materials by a specialist himself in accordance with his information needs.
To achieve the called technical result in each of three first embodiments as well as in the mentioned known method, it is necessary to detect the presence of data in the processed materials which can be described on their content by elements of the characteristic of user's information needs (“user's profile”), to fix the presence of such data and corresponding thereto elements of the characteristic of “user's profile”, and to use these elements and combinations thereof in providing the content of the processed materials to the user.
In contrast with the known method in the first, second and third suggested methods, the processing is performed in an interactive mode with demonstrating sequently separate sense fragments on which information materials being processed are subdivided, in a form corresponding to types of the fragments, with demonstrating the elements of the characteristic of “user's profile” in visual or audiovisual forms.
In case of detection of sense relationship between the content of a definite fragment and some or other from the mentioned elements, the presence of such a relationship is fixed by generating an individual feature for each of the elements with which the relationship is detected.
Further, in the first suggested method with detecting different order relationship of the given fragment with different elements of the characteristic of “user's profile”, the features of relation of these elements to different tiers should be generated in accordance with the number of the detected grades of the relationship. After that, an image of local structure, i.e. a structure relating to the given fragment of the structure, is being formed and representing a marked connected undirected graph, to nodes of which elements of the characteristic of “user's profile”, for which the features of the presence of relations with the content of the given fragment were formed, are put in accordance.
This graph will be fully connected, if no one of the mentioned elements of the characteristic of “user's profile” is dominating that is all of them are considered as of the same tier. If features indicating that elements of the characteristic of “user's profile” belong to different tiers in accordance with detected differences in the density of their relationship with the content of the given fragment, in such a case, each of higher nodes is connected with all lower ones. Moreover, in case of the presence more than one node at the highest tier, each of such nodes is also connected with others and with the same tier.
According to the second suggested method after fixing the presence of the mentioned relationship and forming an individual feature for each of elements of the characteristic of “user's profile” with which the relationship of the content of the given fragment is detected, the demonstration of structures of the connected undirected graph from a fixed set of versions is being performed in visual or audiovisual forms. To nodes of the graph, elements of the characteristic of “user's profile”, for which the features of the presence of relations with the content of the given fragment were formed, are put in accordance. At least one structure of the set is characterized by the presence of a dominating element. Here, as a local structure, i.e. relating to the given fragment, one from the set which is most fitted to this fragment is selected, and as features of the selected structure, the collection of material factors identifying elements of the characteristic of “user's profile” corresponding nodes of this structure and the presence of edges connecting thereof are fixed.
According to the third suggested method after fixing the presence of the mentioned relationship and forming an individual feature for each element of the characteristic of “user's profile”, with the presence of relationship of the content of the give
Bevz Marina Alekseevna
Ivanov Vasiliy Nikolayevich
Kornev Igor Mikhailovich
Lakayev Anatoliy Semenovich
Yakovenko Vitaliy Anatolyevich
Black Thomas
Coby Frantz
Gosudar-Stvenny Nauchno- Tekhnichesky Tsentr Giperinformatsionny
Lackenbach Siegel Marzullo Aronson & Greenspan P.C.
LandOfFree
Method for automatic processing of information materials for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for automatic processing of information materials for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for automatic processing of information materials for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2579661