Filters








1,566 Hits in 5.6 sec

Latent Personality Traits Assessment From Social Network Activity Using Contextual Language Embedding

Pavan Kumar K. N., Marina L. Gavrilova
2021 IEEE Transactions on Computational Social Systems  
The developed system outperforms the state-of-the-art research by reliably estimating the user's latent personality traits while using 50 or fewer tweets per user.  ...  Users reveal aspects of their personality via the content they share with their social media followers and through the patterns in their interactions on online networking platforms.  ...  It incorporates frequency (F) of words and phrases (via TF-IDF), co-occurrence (C) statistics of words from a massive corpus (from GloVe), and contextual (C) meaning of a word (from USE).  ... 
doi:10.1109/tcss.2021.3108810 fatcat:ni2fea2hkndjpjxkl6f5tl7g4m

What's ur Type? Contextualized Classification of User Types in Marijuana-Related Communications Using Compositional Multiview Embedding

Ugur Kursuncu, Manas Gaur, Usha Lokala, Anurag Illendula, Krishnaprasad Thirunarayan, Raminta Daniulaityte, Amit Sheth, I. Budak Arpinar
2018 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI)  
With 93% of pro-marijuana population in US favoring legalization of medical marijuana 1 , high expectations of a greater return for Marijuana stocks 2 , and public actively sharing information about medical  ...  We developed a comprehensive approach to classifying users by their types on Twitter through contextualization of their marijuana-related conversations.  ...  Word Embedding Model A word embedding model created using word2vec can learn a rich low dimensional representation of words in a tweet corpus.  ... 
doi:10.1109/wi.2018.00-50 dblp:conf/webi/KursuncuGLITDSA18 fatcat:uppbpagtajc7bb47223r264wae

Contextual Multi-View Query Learning for Short Text Classification in User-Generated Data [article]

Payam Karisani, Negin Karisani, Li Xiong
2021 arXiv   pre-print
COCOBA employs the context of user postings to construct two views. Then it uses the distribution of the representations in each view to detect the regions that are assigned to the opposite classes.  ...  ., for the early detection of outbreaks or for extracting personal observations--often suffers from the lack of enough training data, short document length, and informal language model.  ...  (Section 3.1): We used neural contextual word embeddings to represent the two Training Test contextual representations discussed in Section 3.1  ... 
arXiv:2112.02611v1 fatcat:t4c63auyqndwrpvx6xs3afeqoq

"What's ur type?" Contextualized Classification of User Types in Marijuana-related Communications using Compositional Multiview Embedding [article]

Ugur Kursuncu, Manas Gaur, Usha Lokala, Anurag Illendula, Krishnaprasad Thirunarayan, Raminta Daniulaityte, Amit Sheth, I. Budak Arpinar
2018 arXiv   pre-print
With 93% of pro-marijuana population in US favoring legalization of medical marijuana, high expectations of a greater return for Marijuana stocks, and public actively sharing information about medical,  ...  We developed a comprehensive approach to classifying users by their types on Twitter through contextualization of their marijuana-related conversations.  ...  The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.  ... 
arXiv:1806.06813v1 fatcat:fc3vokveyvdpnpjc2ejcaihu4a

"When they say weed causes depression, but it's your fav antidepressant": Knowledge-aware Attention Framework for Relationship Extraction [article]

Shweta Yadav, Usha Lokala, Raminta Daniulaityte, Krishnaprasad Thirunarayan, Francois Lamy, Amit Sheth
2020 arXiv   pre-print
With the increasing legalization of medical and recreational use of cannabis, more research is needed to understand the association between depression and consumer behavior related to cannabis consumption  ...  Experimental results show that inclusion of the knowledge-aware attentive representation in association with BERT can extract the cannabis-depression relationship with better coverage in comparison to  ...  All findings and opinions are of authors and not sponsors.  ... 
arXiv:2009.10155v1 fatcat:hoq7jnuod5bbnjmp7e4qv5r3bm

Tweedle: Sensitivity Check in Health-related Social Short Texts based on Regret Theory

R Geetha, S Karthika, N Pavithra, V Preethi
2019 Procedia Computer Science  
Every day, millions of Twitteraties tweet something personal or impersonal to express their emotions and valuable knowledge.  ...  Every day, millions of Twitteraties tweet something personal or impersonal to express their emotions and valuable knowledge.  ...  Though there are many privacy settings available on social networking sites, the users must be aware of using it in the right way.  ... 
doi:10.1016/j.procs.2020.01.062 fatcat:7lov77shqze63jdrzpv5jsvgoy

SmokEng: Towards Fine-grained Classification of Tobacco-related Social Media Text [article]

Kartikey Pant, Venkata Himakar Yanamandra, Alok Debnath, Radhika Mamidi
2019 arXiv   pre-print
In this paper, we create a dataset of 3144 tweets, which are selected based on the presence of colloquial slang related to smoking and analyze it based on the semantics of the tweet.  ...  Each class is created and annotated based on the content of the tweets such that further hierarchical methods can be easily applied.  ...  One of the more recent contextualized word representation is BERT (Devlin et al., 2019) .  ... 
arXiv:1910.05598v1 fatcat:vqb72ttn7rgppf7k7frgsm57za

Using Contextual Information to Improve Blood Glucose Prediction [article]

Mohammad Akbari, Rumi Chunara
2019 arXiv   pre-print
Given a robust evaluation across two blood glucose datasets with different forms of contextual information, we conclude that multi-signal Gaussian Processes can improve blood glucose prediction by using  ...  Therefore, here we propose a Gaussian Process model to both address these data challenges and combine blood glucose and latent feature representations of contextual data for a novel multi-signal blood  ...  Table 5 presents six distinct attributes we extracted and used in our prediction task. proportion of tweets with hashtags u 5 avg. number of tweets/day u 6 total number of tweets Content-centric Features  ... 
arXiv:1909.01735v1 fatcat:jt27zxuwi5d6zjw7btqvkjlkjy

Automatic Breast Cancer Survivor Detection from Social Media for Studying Latent Factors Affecting Treatment Success [article]

Abeed Sarker, Mohammed Ali Al-Garadi, Yuan-Chi Yang, Sahithi Lakamana, Jie Lin, Sabrina Li, Angel Xie, Whitney Hogg-Bremer, Mylin Torres, Imon Banerjee, Abeed Sarker
2020 medRxiv   pre-print
manually-annotated data (n=5019) for distinguishing firsthand self-reports of breast cancer from other tweets.  ...  A classifier based on bidirectional encoder representations from transformers (BERT) showed human-like performance and achieved F1-score of 0.857 (inter-annotator agreement: 0.845; Cohen's kappa) for the  ...  However, such representations do not capture contextual differences in the meanings of words.  ... 
doi:10.1101/2020.05.17.20104778 fatcat:vwncebqdvbbh3cu25m2v2atnq4

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

2018 JAMIA Journal of the American Medical Informatics Association  
and normalization of health-related text from social media.  ...  We executed the Social Media Mining for Health (SMM4H) 2017 shared tasks to enable the community-driven development and large-scale evaluation of automatic text processing methods for the classification  ...  Domain expertize for the University of Manchester team was provided by Professor William G. Dixon, Director of the Arthritis Research U.K. Centre for Epidemiology.  ... 
doi:10.1093/jamia/ocy114 pmid:30272184 pmcid:PMC6188524 fatcat:xcnb65ojmvbh7e55ylxhsnusrm

"When they say weed causes depression, but it's your fav antidepressant": Knowledge-aware attention framework for relationship extraction

Shweta Yadav, Usha Lokala, Raminta Daniulaityte, Krishnaprasad Thirunarayan, Francois Lamy, Amit Sheth, Robert Hoehndorf
2021 PLoS ONE  
With the increasing legalization of medical and recreational use of cannabis, more research is needed to understand the association between depression and consumer behavior related to cannabis consumption  ...  Experimental results show that inclusion of the knowledge-aware attentive representation in association with BERT can extract the cannabis-depression relationship with better coverage in comparison to  ...  We obtained the tweet representation via BERT model as follows: B ¼ 1 n X n j¼1 H LÀ 1 b ½j; :� ð10Þ In our experiments, the representation obtained from the second last (L − 1) Transformer layer achieved  ... 
doi:10.1371/journal.pone.0248299 pmid:33764983 fatcat:awylmujzffc5tjxvhy75bnfbje

Text Classification Models for the Automatic Detection of Nonmedical Prescription Medication Use from Social Media [article]

Mohammed Ali Al-Garadi, Yuan-Chi Yang, Haitao Cai, Yucheng Ruan, Karen O'Connor, Graciela Gonzalez-Hernandez, Jeanmarie Perrone, Abeed Sarker
2020 medRxiv   pre-print
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  ...  We illustrate, via experimentation using differing training set sizes, that the transformer-based models are more stable and require less annotated data compared to the other models.  ...  The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.  ... 
doi:10.1101/2020.04.13.20064089 fatcat:iei7ywd3lzc3dejklcyhfrvoty

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

Mohammed Ali Al-Garadi, Yuan-Chi Yang, Haitao Cai, Yucheng Ruan, Karen O'Connor, Gonzalez-Hernandez Graciela, Jeanmarie Perrone, Abeed Sarker
2021 BMC Medical Informatics and Decision Making  
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  ...  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.  ...  dictionary with 855 clusters of generalized representations of words learned from medication-related chatter collected on Twitter).  ... 
doi:10.1186/s12911-021-01394-0 pmid:33499852 fatcat:peoffb3hojbi5chkzkvt2klcjy

Using Twitter Data to Monitor Natural Disaster Social Dynamics: A Recurrent Neural Network Approach with Word Embeddings and Kernel Density Estimation

Aldo Hernandez-Suarez, Gabriel Sanchez-Perez, Karina Toscano-Medina, Hector Perez-Meana, Jose Portillo-Portillo, Victor Sanchez and Luis, Luis Javier García Villalba
2019 Sensors  
The proposed approach is carried out by transforming tokenized tweets into word embeddings: a rich linguistic and contextual vector representation of textual corpora.  ...  at which a person was last seen.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/s19071746 fatcat:lrgooqmx2fhprjotv7a3w6vhjy

Fine-Grained Sentiment Analysis of Arabic COVID-19 Tweets Using BERT-Based Transformers and Dynamically Weighted Loss Function

Nora Alturayeif, Hamzah Luqman
2021 Applied Sciences  
We also show how the textual representation of emojis can boost the performance of sentiment analysis.  ...  multi-dialect Arabic tweets with an F1-Micro score of 0.72.  ...  Acknowledgments: The authors would like to acknowledge the support provided by King Fahd University of Petroleum and Minerals (KFUPM) during this work.  ... 
doi:10.3390/app112210694 fatcat:2t7fqfmzr5ezvhr2ytcpynkliy
« Previous Showing results 1 — 15 out of 1,566 results