Literature DB >> 33499852

Text classification models for the automatic detection of nonmedical prescription medication use from social media.

Mohammed Ali Al-Garadi1, Yuan-Chi Yang2, Haitao Cai3, Yucheng Ruan4, Karen O'Connor3, Gonzalez-Hernandez Graciela3, Jeanmarie Perrone5, Abeed Sarker2,6.   

Abstract

BACKGROUND: Prescription medication (PM) misuse/abuse has emerged as a national crisis in the United States, and social media has been suggested as a potential resource for performing active monitoring. However, automating a social media-based monitoring system is challenging-requiring advanced natural language processing (NLP) and machine learning methods. In this paper, we describe the development and evaluation of automatic text classification models for detecting self-reports of PM abuse from Twitter.
METHODS: We experimented with state-of-the-art bi-directional transformer-based language models, which utilize tweet-level representations that enable transfer learning (e.g., BERT, RoBERTa, XLNet, AlBERT, and DistilBERT), proposed fusion-based approaches, and compared the developed models with several traditional machine learning, including deep learning, approaches. Using a public dataset, we evaluated the performances of the classifiers on their abilities to classify the non-majority "abuse/misuse" class.
RESULTS: Our proposed fusion-based model performs significantly better than the best traditional model (F1-score [95% CI]: 0.67 [0.64-0.69] vs. 0.45 [0.42-0.48]). We illustrate, via experimentation using varying training set sizes, that the transformer-based models are more stable and require less annotated data compared to the other models. The significant improvements achieved by our best-performing classification model over past approaches makes it suitable for automated continuous monitoring of nonmedical PM use from Twitter.
CONCLUSIONS: BERT, BERT-like and fusion-based models outperform traditional machine learning and deep learning models, achieving substantial improvements over many years of past research on the topic of prescription medication misuse/abuse classification from social media, which had been shown to be a complex task due to the unique ways in which information about nonmedical use is presented. Several challenges associated with the lack of context and the nature of social media language need to be overcome to further improve BERT and BERT-like models. These experimental driven challenges are represented as potential future research directions.

Entities:  

Keywords:  Machine learning; Natural language processing; Prescription medication misuse; Social media

Mesh:

Substances:

Year:  2021        PMID: 33499852      PMCID: PMC7835447          DOI: 10.1186/s12911-021-01394-0

Source DB:  PubMed          Journal:  BMC Med Inform Decis Mak        ISSN: 1472-6947            Impact factor:   2.796


  20 in total

1.  Filtering big data from social media--Building an early warning system for adverse drug reactions.

Authors:  Ming Yang; Melody Kiang; Wei Shang
Journal:  J Biomed Inform       Date:  2015-02-14       Impact factor: 6.317

2.  Epidemiology from Tweets: Estimating Misuse of Prescription Opioids in the USA from Social Media.

Authors:  Michael Chary; Nicholas Genes; Christophe Giraud-Carrier; Carl Hanson; Lewis S Nelson; Alex F Manini
Journal:  J Med Toxicol       Date:  2017-08-22

3.  Characteristics of state prescription drug monitoring programs: a state-by-state survey.

Authors:  A Travis Manasco; Christopher Griggs; Rebecca Leeds; Breanne K Langlois; Alan H Breaud; Patricia M Mitchell; Scott G Weiner
Journal:  Pharmacoepidemiol Drug Saf       Date:  2016-04-08       Impact factor: 2.890

4.  Candyflipping and Other Combinations: Identifying Drug-Drug Combinations from an Online Forum.

Authors:  Michael Chary; David Yi; Alex F Manini
Journal:  Front Psychiatry       Date:  2018-04-30       Impact factor: 4.157

5.  Drug Use in the Twittersphere: A Qualitative Contextual Analysis of Tweets About Prescription Drugs.

Authors:  Lukas Shutler; Lewis S Nelson; Ian Portelli; Courtney Blachford; Jeanmarie Perrone
Journal:  J Addict Dis       Date:  2015

6.  Data and systems for medication-related text classification and concept normalization from Twitter: insights from the Social Media Mining for Health (SMM4H)-2017 shared task.

Authors:  Abeed Sarker; Maksim Belousov; Jasper Friedrichs; Kai Hakala; Svetlana Kiritchenko; Farrokh Mehryary; Sifei Han; Tung Tran; Anthony Rios; Ramakanth Kavuluru; Berry de Bruijn; Filip Ginter; Debanjan Mahata; Saif M Mohammad; Goran Nenadic; Graciela Gonzalez-Hernandez
Journal:  J Am Med Inform Assoc       Date:  2018-10-01       Impact factor: 4.497

7.  Promoting Reproducible Research for Characterizing Nonmedical Use of Medications Through Data Annotation: Description of a Twitter Corpus and Guidelines.

Authors:  Karen O'Connor; Abeed Sarker; Jeanmarie Perrone; Graciela Gonzalez Hernandez
Journal:  J Med Internet Res       Date:  2020-02-26       Impact factor: 5.428

8.  Tweaking and tweeting: exploring Twitter for nonmedical use of a psychostimulant drug (Adderall) among college students.

Authors:  Carl L Hanson; Scott H Burton; Christophe Giraud-Carrier; Josh H West; Michael D Barnes; Bret Hansen
Journal:  J Med Internet Res       Date:  2013-04-17       Impact factor: 5.428

9.  The Canary in the Coal Mine Tweets: Social Media Reveals Public Perceptions of Non-Medical Use of Opioids.

Authors:  Brian Chan; Andrea Lopez; Urmimala Sarkar
Journal:  PLoS One       Date:  2015-08-07       Impact factor: 3.240

10.  Social Media Mining for Toxicovigilance: Automatic Monitoring of Prescription Medication Abuse from Twitter.

Authors:  Abeed Sarker; Karen O'Connor; Rachel Ginn; Matthew Scotch; Karen Smith; Dan Malone; Graciela Gonzalez
Journal:  Drug Saf       Date:  2016-03       Impact factor: 5.606

View more
  7 in total

1.  Signals of increasing co-use of stimulants and opioids from online drug forum data.

Authors:  Abeed Sarker; Mohammed Ali Al-Garadi; Yao Ge; Nisha Nataraj; Christopher M Jones; Steven A Sumner
Journal:  Harm Reduct J       Date:  2022-05-25

2.  The Impact of COVID-19 on Consumers' Psychological Behavior Based on Data Mining for Online User Comments in the Catering Industry in China.

Authors:  Chenyu Zhang; Jiayue Jiang; Hong Jin; Tinggui Chen
Journal:  Int J Environ Res Public Health       Date:  2021-04-15       Impact factor: 3.390

3.  Clinical Decision Support System Based on Hybrid Knowledge Modeling: A Case Study of Chronic Kidney Disease-Mineral and Bone Disorder Treatment.

Authors:  Syed Imran Ali; Su Woong Jung; Hafiz Syed Muhammad Bilal; Sang-Ho Lee; Jamil Hussain; Muhammad Afzal; Maqbool Hussain; Taqdir Ali; Taechoong Chung; Sungyoung Lee
Journal:  Int J Environ Res Public Health       Date:  2021-12-26       Impact factor: 3.390

4.  A Fine-Tuned BERT-Based Transfer Learning Approach for Text Classification.

Authors:  Rukhma Qasim; Waqas Haider Bangyal; Mohammed A Alqarni; Abdulwahab Ali Almazroi
Journal:  J Healthc Eng       Date:  2022-01-07       Impact factor: 2.682

5.  Comparison of Pretraining Models and Strategies for Health-Related Social Media Text Classification.

Authors:  Yuting Guo; Yao Ge; Yuan-Chi Yang; Mohammed Ali Al-Garadi; Abeed Sarker
Journal:  Healthcare (Basel)       Date:  2022-08-05

6.  Detection of Depression Severity Using Bengali Social Media Posts on Mental Health: Study Using Natural Language Processing Techniques.

Authors:  Muhammad Khubayeeb Kabir; Maisha Islam; Anika Nahian Binte Kabir; Adiba Haque; Md Khalilur Rhaman
Journal:  JMIR Form Res       Date:  2022-09-28

7.  Automatic gender detection in Twitter profiles for health-related cohort studies.

Authors:  Yuan-Chi Yang; Mohammed Ali Al-Garadi; Jennifer S Love; Jeanmarie Perrone; Abeed Sarker
Journal:  JAMIA Open       Date:  2021-06-23
  7 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.