Time Matters: Time-Aware LSTMs for Predictive Business Process Monitoring

نویسندگان

چکیده

Predictive business process monitoring (PBPM) aims to predict future behavior during ongoing executions based on event log data. Especially, techniques for the next activity and timestamp prediction can help improve performance of operational processes. Recently, many PBPM solutions deep learning were proposed by researchers. Due sequential nature data, a common choice is apply recurrent neural networks with long short-term memory (LSTM) cells. We argue, that elapsed time between events informative. However, current mainly use “vanilla” LSTM cells hand-crafted time-related control flow features. To better model dependencies events, we propose new technique time-aware (T-LSTM) T-LSTM incorporate consecutive inherently adjust cell memory. Furthermore, introduce cost-sensitive account class imbalance in logs. Our experiments publicly available benchmark logs indicate effectiveness introduced techniques.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Survey on Time-aware Business Process Modeling

One key perspective when dealing with Business Process Management (BPM) is time. All business experts agree upon the fact that time is a key resource for each business process within an organisation. Indeed, time managing is an effective cost and error reduction strategy and thus ensures profit maximization for the organisation. As a result, business managers, researchers, and academicians in m...

متن کامل

Real-time business process monitoring using formal concept analysis

Purpose To help industrial managers monitor and analyze critical performance indicators in real-time during the execution of business processes, we propose a visualization technique using an extended formal concept analysis (FCA). The proposed approach monitors the current progress of ongoing processes and periodically predicts their probable routes and performances. Design/methodology/approach...

متن کامل

A General Framework for Predictive Business Process Monitoring

As organizations gain awareness of the potential business value locked in their process execution event logs, “evidence-based” business process management (BPM) becomes a common tool for process analysts. In contrast to traditional process monitoring techniques which are typically performed using data from running process instances only, predictive evidence-based BPM methods tap also into histo...

متن کامل

Verifying Controllability of Time-Aware Business Processes

We present an operational semantics for time-aware business processes, that is, processes modeling the execution of business activities, whose durations are subject to linear constraints over the integers. We assume that some of the durations are controllable, that is, they can be determined by the organization that enacts the process, while others are uncontrollable, that is, they are determin...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Lecture notes in business information processing

سال: 2021

ISSN: ['1865-1348', '1865-1356']

DOI: https://doi.org/10.1007/978-3-030-72693-5_9