Automated retraining of machine learning models

Akanksha Kavikondala, Vivek Muppalla, K. Krishna Prakasha, Vasundhara Acharya

Research output: Contribution to journalArticle

Abstract

Data is the most crucial component of a successful ML system. Once a machine learning model is developed, it gets obsolete over time due to presence of new input data being generated every second. In order to keep our predictions accurate we need to find a way to keep our models up to date. Our research work involves finding a mechanism which can retrain the model with new data automatically. This research also involves exploring the possibilities of automating machine learning processes. We started this project by training and testing our model using conventional machine learning methods. The outcome was then compared with the outcome of those experiments conducted using the AutoML methods like TPOT. This helped us in finding an efficient technique to retrain our models. These techniques can be used in areas where people do not deal with the actual working of a ML model but only require the outputs of ML processes.

Original languageEnglish
Pages (from-to)445-452
Number of pages8
JournalInternational Journal of Innovative Technology and Exploring Engineering
Volume8
Issue number12
DOIs
Publication statusPublished - 01-10-2019
Externally publishedYes

Fingerprint

Learning systems
Testing
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Civil and Structural Engineering
  • Mechanics of Materials
  • Electrical and Electronic Engineering

Cite this

Kavikondala, Akanksha ; Muppalla, Vivek ; Krishna Prakasha, K. ; Acharya, Vasundhara. / Automated retraining of machine learning models. In: International Journal of Innovative Technology and Exploring Engineering. 2019 ; Vol. 8, No. 12. pp. 445-452.
@article{80308ca2bed141f9b48874e9bf203a4b,
title = "Automated retraining of machine learning models",
abstract = "Data is the most crucial component of a successful ML system. Once a machine learning model is developed, it gets obsolete over time due to presence of new input data being generated every second. In order to keep our predictions accurate we need to find a way to keep our models up to date. Our research work involves finding a mechanism which can retrain the model with new data automatically. This research also involves exploring the possibilities of automating machine learning processes. We started this project by training and testing our model using conventional machine learning methods. The outcome was then compared with the outcome of those experiments conducted using the AutoML methods like TPOT. This helped us in finding an efficient technique to retrain our models. These techniques can be used in areas where people do not deal with the actual working of a ML model but only require the outputs of ML processes.",
author = "Akanksha Kavikondala and Vivek Muppalla and {Krishna Prakasha}, K. and Vasundhara Acharya",
year = "2019",
month = "10",
day = "1",
doi = "10.35940/ijitee.L3322.1081219",
language = "English",
volume = "8",
pages = "445--452",
journal = "International Journal of Innovative Technology and Exploring Engineering",
issn = "2278-3075",
publisher = "Blue Eyes Intelligence Engineering and Sciences Publication",
number = "12",

}

Automated retraining of machine learning models. / Kavikondala, Akanksha; Muppalla, Vivek; Krishna Prakasha, K.; Acharya, Vasundhara.

In: International Journal of Innovative Technology and Exploring Engineering, Vol. 8, No. 12, 01.10.2019, p. 445-452.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Automated retraining of machine learning models

AU - Kavikondala, Akanksha

AU - Muppalla, Vivek

AU - Krishna Prakasha, K.

AU - Acharya, Vasundhara

PY - 2019/10/1

Y1 - 2019/10/1

N2 - Data is the most crucial component of a successful ML system. Once a machine learning model is developed, it gets obsolete over time due to presence of new input data being generated every second. In order to keep our predictions accurate we need to find a way to keep our models up to date. Our research work involves finding a mechanism which can retrain the model with new data automatically. This research also involves exploring the possibilities of automating machine learning processes. We started this project by training and testing our model using conventional machine learning methods. The outcome was then compared with the outcome of those experiments conducted using the AutoML methods like TPOT. This helped us in finding an efficient technique to retrain our models. These techniques can be used in areas where people do not deal with the actual working of a ML model but only require the outputs of ML processes.

AB - Data is the most crucial component of a successful ML system. Once a machine learning model is developed, it gets obsolete over time due to presence of new input data being generated every second. In order to keep our predictions accurate we need to find a way to keep our models up to date. Our research work involves finding a mechanism which can retrain the model with new data automatically. This research also involves exploring the possibilities of automating machine learning processes. We started this project by training and testing our model using conventional machine learning methods. The outcome was then compared with the outcome of those experiments conducted using the AutoML methods like TPOT. This helped us in finding an efficient technique to retrain our models. These techniques can be used in areas where people do not deal with the actual working of a ML model but only require the outputs of ML processes.

UR - http://www.scopus.com/inward/record.url?scp=85073983064&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073983064&partnerID=8YFLogxK

U2 - 10.35940/ijitee.L3322.1081219

DO - 10.35940/ijitee.L3322.1081219

M3 - Article

AN - SCOPUS:85073983064

VL - 8

SP - 445

EP - 452

JO - International Journal of Innovative Technology and Exploring Engineering

JF - International Journal of Innovative Technology and Exploring Engineering

SN - 2278-3075

IS - 12

ER -