LEARNING APPARATUS, DETERMINING APPARATUS, LEARNING METHOD, DETERMINING METHOD, LEARNING PROGRAM AND DETERMINING PROGRAM

• Main Author: AMEMIYA SATOSHI
• Format: Patent
• Language: English; Japanese
• Published: 12.03.2020
• Subjects: CALCULATING; CHAMBERS PROVIDED WITH MANIPULATION DEVICES; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; COUNTING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; HAND TOOLS; MANIPULATORS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PERFORMING OPERATIONS; PHYSICS; PORTABLE POWER-DRIVEN TOOLS; REGULATING; TRANSPORTING

To provide a learning apparatus, determining apparatus, learning method, determining method, learning program and determining program capable of improving the accuracy of determination.SOLUTION: A learning apparatus comprises: a probability distribution preparing section that prepares probability distributions respectively on appliances from for-learning sensor data acquired from sensors provided at a plurality of appliances; a difference-degree calculating section that calculates a difference degree between the probability distributions; a probability model preparing section that prepares a probability model by combining a plurality of probability distributions in which the difference degree is fallen smaller than a threshold into one probability distribution and summing up multiplications of a coefficient on the probability distributions; and a criterion preparing section that prepares an abnormality determining criterion from the probability model.SELECTED DRAWING: Figure 6 【課題】 判定精度を向上させることができる学習装置、判定装置、学習方法、判定方法、学習プログラムおよび判定プログラムを提供する。【解決手段】 学習装置は、複数の機器に設けられたセンサから取得した学習用センサデータから、各機器についてそれぞれ確率分布を作成する確率分布作成部と、各確率分布同士の相違度を算出する相違度算出部と、前記相違度が閾値未満となる複数の確率分布を１つの確率分布に合成し、各確率分布に係数を掛け合わせたうえで足し合わせることで確率モデルを作成する確率モデル作成部と、前記確率モデルから異常判定用基準を作成する基準作成部と、を備える。【選択図】 図６