基于深度半非負(fù)矩陣分解的醫(yī)保異常行為檢測

打開文本圖片集

摘  要：醫(yī)療保險是一種提供因疾病或傷害而支付的保險福利，檢測醫(yī)保數(shù)據(jù)和嚴(yán)厲打擊欺詐騙保行為是維持醫(yī)保系統(tǒng)健康發(fā)展的必要手段。該文利用機器學(xué)習(xí)技術(shù)，收集醫(yī)療保險大數(shù)據(jù)，通過對原始數(shù)據(jù)進(jìn)行降維和聚類分析，對醫(yī)保數(shù)據(jù)行為進(jìn)行異常檢測，最終對是否存在欺詐騙保行為作出正確決策。

關(guān)鍵詞：醫(yī)療保險;欺詐騙保;機器學(xué)習(xí);異常檢測;降維分析

中圖分類號：TP311 文獻(xiàn)標(biāo)志碼：A 文章編號：2095-2945（2022）25-0090-04

Abstract： Medical insurance is a kind of insurance benefits paid for illness or injury. Detecting medical insurance data and cracking down on insurance fraud are necessary means to maintain the healthy development of medical insurance system.In this paper， machine learning technology is used to big data about collect medical insurance， and abnormal detection of medical insurance data behavior is conducted through downgrading and clustering analysis of original data， so as to make correct decisions on whether there is fraud or not.

Keywords： medical insurance; fraud insurance; machine learning; anomaly detection; dimension reduction analysis

構(gòu)建良好的社會保障體系是當(dāng)今時代保障社會生活穩(wěn)定發(fā)展，不斷滿足人民日益增長的美好生活需要的必要舉措。（剩余6966字）