Explaining and Predicting Abnormal Expenses at Large Scale Using Knowledge Graph Based Reasoning

Abstract

Global business travel spend topped record-breaking $1.2 Trillion USD in 2015, and will reach $1.6 Trillion by 2020 according to the Global Business Travel Association, the world’s premier business travel and meetings trade organization. Existing expenses systems are designed for reporting expenses, their type and amount over pre-defined views such as time period, service or employee group. However such systems do not aim at systematically detecting abnormal expenses, and more importantly explaining their causes. Therefore deriving any actionable insight for optimizing spending and saving from their analysis is time-consuming, cumbersome and often impossible. Towards this challenge we present AIFS, a system designed for expenses business owner and auditors. Our system is manipulating and combining semantic web and machine learning technologies for (i) identifying, (ii) explaining and (iii) predicting abnormal expenses claim by employees of large organisations. Our prototype of semantics-aware employee expenses analytics and reasoning, experimented with 191; 346 unique Accenture employees in 2015, has demonstrated scalability and accuracy for the tasks of explaining and predicting abnormal expenses.