EP0616291A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
A method of configuring a neural network and a diagnosis/recognition system using the same
Assignee: Hitachi Ltd
•
Filing: 1994-02-01
•
Granted: —
•
Expires: 2014-02-01
Abstract
Abstract In a recognition/diagnosis method, a relation between total sums of inputs of respective cases and teacher data is listed in the order of magnitude of the total sums of inputs. Based on the value of the teacher data for the input having the maximum total sum and the number of times of change of teacher data in the table are considered, a configuration of a neural network (the number of hidden layers and the number of neurons thereof) is determined. Coupling weights are analytically calculated based on the table.
Inventors
- 👤 Isao Takahashi
- 👤 Fumihiro Endo
- 👤 Tokio Yamagiwa
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-29 21:59
CPC Classifications
G; G06; G06N; G06N3/00; G06N3/02; G06N3/08; G06N3/04; G06N3/0499; G06N3/09
Claims Analysis
Total Claims: 19
Claim 1
Independent Claim
104 words
A recognition or diagnosis method using a neural network having an input layer, at least a hidden layer and an output layer comprising the steps of: classifying learning data inputted to said input layer based on features thereof; inputting learning data for each class; calculating a total sum for each case; correlating corresponding teacher data to the total sum in the order of magnitude for each output of said output layer; and determining coupling weights based on the teacher data corresponding to a case having a maximum total sum, total sums before and after the change of the teacher data and the teacher data.
Claim 2
Dependent Claim
116 words
A recognition or diagnosis method using a neural network having an input layer, at least a hidden layer and an output layer comprising the steps of: inputting input data for each value for each case based on an upper limit and a lower limit of the input data of each case inputted to said input layer; calculating a total sum for each case; correlating corresponding teacher data to the total sum in the order of magnitude for each output of said output layer; and determining coupling weights based on the teacher data corresponding to a case having a maximum total sum, total sums before and after the change of the teacher data and the teacher data.
Claim 3
Dependent Claim
34 words
A recognition or diagnosis method according to Claim 2 wherein the input data inputted to said input layer are products of the upper limit and the lower limit multiplied by the same coupling weight.
Claim 4
Dependent Claim
151 words
A recognition or diagnosis method using a neural network having an input layer, at least a hidden layer and an output layer comprising the steps of: selecting coupling weights such that an input sum determined for a case other than a particular case inputted to said input layer does not fall between two input sums determined based on an upper limit and a lower limit of input data of said particular case and determining the input sum; inputting said input sum as input data; calculating an input sum for each case; calculating a total sum for each case; correlating corresponding teacher data to the input sum in the order of magnitude for each output of said output layer; and determining coupling weights based on the teacher data corresponding to a case having a maximum total sum, total sums before and after the change of the teacher data and the teacher data.
Claim 5
Dependent Claim
34 words
A recognition or diagnosis method according to Claim 4 wherein the teacher data of said particular case and the teacher data of a case other than said particular case are different from each other.
Claim 6
Dependent Claim
56 words
A recognition or diagnosis method according to Claim 1 wherein two hidden layers are provided and the coupling weights are selected such that when the teacher data corresponding to the case having the maximum total sum is 0, the second hidden layer functions as an AND circuit and said output layer functions as an OR circuit.
Claim 7
Dependent Claim
56 words
A recognition or diagnosis method according to Claim 1 wherein two hidden layers are provided and the coupling weights are selected such that when the teacher data corresponding to the case having the maximum total sum is 1, the second hidden layer functions as an OR circuit and said output layer functions as an AND circuit.
Claim 8
Dependent Claim
33 words
A recognition or diagnosis method according to Claim 1 wherein said data are used by reciprocals thereof for each case and a reversal operation in the magnitude of the input data is executed.
Claim 9
Dependent Claim
31 words
A recognition or diagnosis method according to Claim 1 wherein a back propagation iterative learning method is used to calculate the coupling weights which are principal parts of the neural network.
Claim 10
Dependent Claim
118 words
A recognition or diagnosis method using a neural network having an input layer, at least a hidden layer and an output layer comprising the steps of: inputting learning data to said input layer for each case; calculating a total sum for each case; correlating corresponding teacher data to the total sum in the order of magnitude for each output of said output layer; determining coupling weights based on the teacher data corresponding to a case having a maximum total sum, total sums before and after the change of the teacher data and the teacher data; and indicating the deviation from a specified range of scatter of the learning data when input diagnosis data is out of the range.
Claim 11
Dependent Claim
129 words
A recognition or diagnosis method using a neural network having an input layer, at least a hidden layer and an output layer comprising the steps of: inputting learning data to said input layer for each case; calculating a total sum for each case; correlating corresponding teacher data to the total sum in the order of magnitude for each output of said output layer; determining coupling weights based on the teacher data corresponding to a case having a maximum total sum, total sums before and after the change of the teacher data and the teacher data; and transferring diagnosis data and the corresponding correct solution as a portion of the learning data after the diagnosis when the correct solution to the diagnosis data is known in inputting the diagnosis data.
Claim 12
Dependent Claim
131 words
A recognition or diagnosis method using a neural network having an input layer, at least a hidden layer and an output layer comprising the steps of: inputting learning data to said input layer for each case; calculating a total sum for each case; correlating corresponding teacher data to the total sum in the order of magnitude for each output of said output layer; determining coupling weights based on the teacher data corresponding to a case having a maximum total sum, total sums before and after the change of the teacher data and the teacher data; and calculating an output for diagnosis data when the diagnosis data is inputted; inferring at least one probable learning data from the output; comparing the diagnosis data with the learning data; and indicating the result thereof.
Claim 13
Dependent Claim
36 words
A recognition or diagnosis method according to Claim 12 wherein said inferred learning data includes a scatter and the diagnosis data is compared with the learning data having the scatter and the result thereof is indicated.
Claim 14
Dependent Claim
34 words
A recognition or diagnosis method according to Claim 1 wherein a first coupling weight to be multiplied to the input data of said input layer is selected to be different from other coupling weights.
Claim 15
Dependent Claim
41 words
A recognition or diagnosis method comprising the steps of: comparing diagnosis data directly with learning data; determining whether the diagnosis data and the learning data of a particular case totally match, partially match or totally mismatch; and indicating the result thereof.
Claim 16
Dependent Claim
46 words
A recognition or diagnosis method comprising the steps of: comparing diagnosis data directly with learning data having a range of scatter; determining whether the diagnosis data is totally within said range, partially within said range or totally out of said range; and indicating the result thereof.
Claim 17
Dependent Claim
57 words
A recognition or diagnosis method comprising the steps of: preparing simultaneous equations by using calculation formulas of inputs to a neuron based on an upper limit and a lower limit of the input data having a range of scatter; solving the simultaneous equations to determine coupling weights; and utilizing a neural network based on the coupling weights.
Claim 18
Dependent Claim
39 words
A recognition or diagnosis method by using a neural network having an input layer, at least a hidden layer and an output layer weighted by coupling weights comprising the step of: self-evaluating teacher data for determining the coupling weights.
Claim 19
Independent Claim
40 words
An apparatus for diagnosing an electrical circuit to be used under a plurality of conditions by using a neural network comprising means for taking the plurality of conditions as learning data to determine coupling weights which form said neural network.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.