Filters








45,379 Hits in 8.0 sec

A Novel Business Process Prediction Model Using a Deep Learning Method

Nijat Mehdiyev, Joerg Evermann, Peter Fettke
2018 Business & Information Systems Engineering  
The paper proposes a multi-stage deep learning approach that formulates the next event prediction problem as a classification problem.  ...  Experiments on a variety of business process log datasets show that the multi-stage deep learning approach provides promising results.  ...  Hence, it is important for classifiers to correctly classify or predict rare events and we therefore ask the following research question: • RQ3: Can process prediction with a multi-stage deep learning  ... 
doi:10.1007/s12599-018-0551-3 fatcat:sbqxbvghkvfbrn4bsxh2v52lgm

POP-ON: Prediction of Process Using One-Way Language Model Based on NLP Approach

Junhyung Moon, Gyuyoung Park, Jongpil Jeong
2021 Applied Sciences  
When solving the next event prediction problem, we use a fully attention-based transformer, which has performed well in recent natural language processing approaches.  ...  It is trained using the proposed deep learning model according to specific pre-processing steps.  ...  Prediction of Process The latest works have evolved from explicit process models to deep learning approaches.  ... 
doi:10.3390/app11020864 fatcat:s3qnt4upirfojetamkx63losim

Time Series Classification using Deep Learning for Process Planning: A Case from the Process Industry

Nijat Mehdiyev, Johannes Lahann, Andreas Emrich, David Enke, Peter Fettke, Peter Loos
2017 Procedia Computer Science  
Recent improvements in the deep learning domain offer opportunities to avoid such an intensive hand-crafted feature engineering which is particularly important for managing the processes based on time-series  ...  In our paper, we propose a framework to extract the features in an unsupervised (or self-supervised) manner using deep learning, particularly stacked LSTM Autoencoder Networks.  ...  Conclusion In the present paper we propose a novel multi-stage deep learning approach for multivariate time series classification problems.  ... 
doi:10.1016/j.procs.2017.09.066 fatcat:igihfvgoejgmdozjacmeyt6lsi

Explainable Artificial Intelligence for Process Mining: A General Overview and Application of a Novel Local Explanation Approach for Predictive Process Monitoring [article]

Nijat Mehdiyev, Peter Fettke
2020 arXiv   pre-print
Consequently, with regard to the theoretical and practical implications of the framework, this study proposes a novel local post-hoc explanation approach for a deep learning classifier that is expected  ...  explainable business process prediction solutions.  ...  In the Section 4, we present the proposed post-hoc local explanation approach for business process prediction scenario and discuss its stages in detail.  ... 
arXiv:2009.02098v2 fatcat:7nf3r24e5redrbnnlayez2t5ke

ORANGE: Outcome-Oriented Predictive Process Monitoring based on Image Encoding and CNNs

Vincenzo Pasquadibisceglie, Annalisa Appice, Giovanna Castellano, Donato Malerba, Giuseppe Modugno
2020 IEEE Access  
CONCLUSION Boosted by the recent application of computer vision approaches in predictive process mining [10] , [30] , we pro- pose a novel deep learning approach for outcome-oriented predictive process  ...  A similar approach is investigated in [29] for business process event prediction, where stacked autoencoders are applied to extract features from the preprocessed business process log data. B.  ...  Her current research interests include data mining with spatio-temporal data, data streams and event logs.  ... 
doi:10.1109/access.2020.3029323 fatcat:2r26nh7u7vgfxlhms2mleuwewe

A systematic literature review on state-of-the-art deep learning methods for process prediction

Dominic A. Neu, Johannes Lahann, Peter Fettke
2021 Artificial Intelligence Review  
Given a sequence of events of an ongoing trace, process prediction allows forecasting upcoming events or performance measurements.  ...  In recent years, multiple process prediction approaches have been proposed, applying different data processing schemes and prediction algorithms.  ...  and comparison of a variety of process prediction datasets deep learning approaches.  ... 
doi:10.1007/s10462-021-09960-8 fatcat:cikw2kxd7vhozeevsrl3qp6p24

Building Interpretable Models for Business Process Prediction using Shared and Specialised Attention Mechanisms [article]

Bemali Wickramanayake, Zhipeng He, Chun Ouyang, Catarina Moreira, Yue Xu, Renuka Sindhgatta
2021 arXiv   pre-print
These predictive models, often built upon deep learning techniques, can be used to make predictions about the future states of business process execution.  ...  Predictive process analytics is a newly emerged discipline dedicated to providing business process intelligence in modern organisations.  ...  key data pre-processing related functions, to prepare the process prefixes into trainable tensors, and gain inspiration for model construction.  ... 
arXiv:2109.01419v1 fatcat:qcvmvi54sfah3bwhdtzwlbqib4

Discovering generative models from event logs: data-driven simulation vs deep learning

Manuel Camargo, Marlon Dumas, Oscar González-Rojas
2021 PeerJ Computer Science  
This paper fills this gap by empirically comparing a data-driven simulation approach with multiple deep learning approaches for building generative business process models.  ...  In the context of business processes, a generative model creates new execution traces from a set of historical traces, also known as an event log.  ...  Reproducibility package for "Discovering Generative Models from Event Logs: Data-driven Simulation vs Deep Learning" (Version 0.1.0). Zenodo. DOI 10.5281/zenodo.4699983.  ... 
doi:10.7717/peerj-cs.577 fatcat:r6szwnrdsfgjfgjdeilxcnlg7q

TFCMA-DRL: Tolerant Flexible Coordinated Multi-Agent Deep Reinforcement Learning for Prediction of Future Stock Price Trends from Multi-Source Data

Chinnasamy Bhuvaneshwari, Kongunadu Arts and Science College, Raman Beena, Kongunadu Arts and Science College
2021 International Journal of Intelligent Engineering and Systems  
Machine learning and deep learning algorithms were the most efficient techniques used for prediction by extracting public opinions and events.  ...  To tackle these issues, the Tolerant Flexible Coordinated Multi-Agent Deep Reinforcement Learning (TFCMA-DRL) model is proposed in this paper.  ...  Author Contributions This work is a contribution of both the authors: Conceptualization, Chinnasamy Bhuvaneshwari  ... 
doi:10.22266/ijies2021.0430.04 fatcat:xrsrb7yjeneorgvsbzlw5tb63y

A systematic literature review on state-of-the-art deep learning methods for process prediction [article]

Dominic A. Neu, Johannes Lahann, Peter Fettke
2021 arXiv   pre-print
Given a sequence of events of an ongoing trace, process prediction allows forecasting upcoming events or performance measurements.  ...  In recent years, multiple process prediction approaches have been proposed, applying different data processing schemes and prediction algorithms.  ...  Science Direct Title, abstract, keywords: "Business Process" AND "Predictive" AND "Deep Learning" 2 Science Direct Title, abstract, keywords: "Business Process" AND "Prediction" AND "Deep Learning" 2 Science  ... 
arXiv:2101.09320v2 fatcat:w33pzhynqbh5zbtv55utqiadqq

ICCUBEA 2019 Table of Contents

2019 2019 5th International Conference On Computing, Communication, Control And Automation (ICCUBEA)  
for Classification of Seizure disorder using EEG Signals 30 Sorghum Leaf Disease Detection and Classification Using Deep Learning 31 Evolutionary Computing Approaches for Solving Multi-Objective  ...  for Autistic People: A survey 22 Steering Angle Prediction in Autonomous Vehicles Using Deep Learning 23 Sentiment Analysis of Telugu data and comparing advanced ensemble techniques using different  ... 
doi:10.1109/iccubea47591.2019.9128707 fatcat:3s3o5v2x3nepxeftk4jd64hjnq

Advanced Customer Activity Prediction based on Deep Hierarchic Encoder-Decoders [article]

Andrei Damian, Laurentiu Piciu, Sergiu Turlea, Nicolae Tapus
2019 arXiv   pre-print
Current state of the art in Deep Learning based recommender systems uses multiple approaches ranging from already classical methods such as the ones based on learning product representation vector, to  ...  In our work we will present a new and innovative architectural approach of applying state-of-the-art hierarchical multi-module encoder-decoder architecture in order to solve several of current state-of-the-art  ...  the first one being the current supervised method and the second one being a finetuning stage using reinforcement learning approach.  ... 
arXiv:1904.07687v4 fatcat:f32a43nyifaq5ky63xd4yxbvvu

Cause vs. Effect in Context-Sensitive Prediction of Business Process Instances [article]

Jens Brunk, Matthias Stierle, Leon Papke, Kate Revoredo, Martin Matzner, Jörg Becker
2020 arXiv   pre-print
Few approaches for tackling this challenge consider a multi-perspective view, where the flow perspective of the process is combined with its surrounding context.  ...  Predicting undesirable events during the execution of a business process instance provides the process participants with an opportunity to intervene and keep the process aligned with its goals.  ...  The fourth author received a grant from Ö sterreichische Akademie der Wissenschaften.  ... 
arXiv:2007.07549v1 fatcat:fqac45hilvbcpdyopaqhggo6ey

A Data-analytics Approach for Enterprise Resilience

Donna Xu, Ivor W. Tsang, Eng K. Chew, Cosimo Siclari, Varun Kaul
2019 IEEE Intelligent Systems  
The recovery process tries to restore the services from negative events and bring them back to business.  ...  The learning process helps to improve the enterprise resilience by learning from the experienced events.  ... 
doi:10.1109/mis.2019.2918092 fatcat:jvk4bgzutfgqrohwtqw7cf7sn4

ARMDN: Associative and Recurrent Mixture Density Networks for eRetail Demand Forecasting [article]

Srayanta Mukherjee, Devashish Shankar, Atin Ghosh, Nilam Tathawadekar, Pramod Kompalli, Sunita Sarawagi, Krishnendu Chaudhury
2018 arXiv   pre-print
We first identify several causal features and use a combination of feature embeddings, MLP and LSTM to represent them. We then model the output density as a learned mixture of Gaussian distributions.  ...  The AR-MDN can be trained end-to-end without the need for additional supervision. We experiment on a dataset of an year's worth of data over tens-of-thousands of products from Flipkart.  ...  Also, motivated by the impressive success of modern deep learning methods on various speech, text, and image processing tasks, we next focused on a neural network model for our demand forecasting problem  ... 
arXiv:1803.03800v2 fatcat:7ziufaneonf3zilcmaca4emgk4
« Previous Showing results 1 — 15 out of 45,379 results