Tweet Authors Research Articles

Unlocking Author Power: On the Exploitation of Auxiliary Author-Retweeter Relations for Predicting Key Retweeters

Retweeting is a powerful driving force in information propagation on microblogging sites. However, identifying the most effective retweeters of a message (called the ”key retweeter prediction” problem) has become a significant research topic. Conventional approaches have addressed this topic from two main aspects: by analyzing either the personal attributes of microblogging users or the structures of user graph networks. However, according to sociological findings, author-retweeter dependencies also play a crucial role in influencing message propagation. In this paper, we propose a novel model to solve the key retweeter prediction problem by incorporating the auxiliary relations between a tweet author and potential retweeters. Without loss of generality, we formulate the relations from four relational factors: status relation, temporal relation, locational relation, and interactive relation. In addition, we propose a novel method, called “Relation-based Learning to Rank (RL2R),” to determine the key retweeters for a given tweet by ranking the potential retweeters in terms of their spreadability. The experimental results show that our method outperforms the state-of-the-art algorithms at top-k retweeter prediction, achieving a significant relative average improvement of 19.7–29.4 percent. These findings provide new insights for understanding user behaviors on social media for key retweeter prediction purposes.

Сеть или текст? Факторы распространения протеста в социальных медиа: теория и анализ данных

Социальные сети способны аккумулировать и концентрировать протестный потенциал, который может затем выплескиваться на улицу. Сетевые связи между индивидами в онлайн-среде родственны связям в физической реальности, и их структура может оказывать влияние на распространение как информации о протесте, так и протестного поведения. Цифровые платформы способны выступать центральными площадками формулирования коллективных целей и идентичностей. Эти положения, сформулированные на основе проведенных ранее исследований, стали отправной точкой настоящей работы, сосредоточенной в ее эмпирической части на венесуэльском сегменте сети Twitter. “Элементарными единицами” протестной активности в Twitter являются отдельные сообщения (твиты), некоторые из которых мультиплицируются (retweet) и достигают аудитории национального масштаба. Именно эти, наиболее популярные твиты играют наибольшую роль в формировании протестного потенциала. Но какие твиты становятся популярными? Те, что написаны авторами, занимающими лучшую позицию в сетевой структуре, или те, которые имеют более яркое, насыщенное содержание? Мы формализовали эту альтернативу, выделив две группы показателей: одну составили индикаторы сетевой позиции автора, другая же характеризует содержание твита. В качестве зависимой переменной выступает количество ретвитов. Проанализировав более 5,7 млн уникальных твитов с использованием современных подходов и методов науки о данных (регрессия с LASSO-регуляризацией, скользящий контроль и др.), мы выявили, что показатели первой группы имеют существенно большее значение. Поскольку в политической онлайн-коммуникации в последние годы значимую роль играют боты (автоматизированные аккаунты, способные, в частности, делать ретвиты), то мы провели дополнительный анализ, “вычистив” из совокупности ретвитов те из них, которые сделаны ботами. Результат оказался в этом случае тем же, что и для исходной выборки: сеть важнее текста. Сетевая позиция автора, ассоциируемая с его популярностью (в особенности – его популярностью среди популярных пользователей), играет гораздо более значимую роль для распространения твита, чем его содержательные характеристики.

Identifying Sleep-Deprived Authors of Tweets: Prospective Study.

BackgroundSocial media data can be explored as a tool to detect sleep deprivation. First-year undergraduate students in their first quarter were invited to wear sleep-tracking devices (Basis; Intel), allow us to follow them on Twitter, and complete weekly surveys regarding their sleep.ObjectiveThis study aimed to determine whether social media data can be used to monitor sleep deprivation.MethodsThe sleep data obtained from the device were utilized to create a tiredness model that aided in labeling the tweets as sleep deprived or not at the time of posting. Labeled data were used to train and test a gated recurrent unit (GRU) neural network as to whether or not study participants were sleep deprived at the time of posting.ResultsResults from the GRU neural network suggest that it is possible to classify the sleep-deprivation status of a tweet’s author with an average area under the curve of 0.68.ConclusionsIt is feasible to use social media to identify students’ sleep deprivation. The results add to the body of research suggesting that social media data should be further explored as a potential source for monitoring health.

A Novel Tweet Recommendation Framework for Twitter

In order to keep them updated users follow various Twitter accounts to get the latest information. As their social network increases it becomes challenging for them to find the relevant content from the massive collection of information. A Twitter user needs to scan a lot of less relevant posts to find the interesting tweets. Important updates may get lost if user is not able to read all the messages. So there is need that the most relevant updates are shown to the user first. Traditionally, the most retweeted tweets are considered popular and are brought forward. In order to improve the attractiveness of the incoming tweets we propose a personalized tweet ranking method based on the trending topics in the user network. A hashtag ranking model is developed to map the tweets into a ranked list of hashtags. The tweets corresponding to those hashtags are then ranked based on the linear weighted model that considers features related to tweet, author of tweet and the user. Finally, conducting a pilot user study we analyze the effectiveness of the proposed framework.

A Hierarchical Model of Nonhomogeneous Poisson Processes for Twitter Retweets

We present a hierarchical model of nonhomogeneous Poisson processes (NHPP) for information diffusion on online social media, in particular Twitter retweets. The retweets of each original tweet are modelled by a NHPP, for which the intensity function is a product of time-decaying components and another component that depends on the follower count of the original tweet author. The latter allows us to explain or predict the ultimate retweet count by a network centrality-related covariate. The inference algorithm enables the Bayes factor to be computed, to facilitate model selection. Finally, the model is applied to the retweet datasets of two hashtags. Supplementary materials for this article, including a standardized description of the materials available for reproducing the work, are available as an online supplement

Exploration of Content and Reach of Physical Therapy-Related Discussion on Twitter.

Many individuals access and share health information on social networking sites. Previous studies have analyzed the social media site Twitter to discern public opinion related to health phenomena such as antibiotics, concussions, and flu. Physical therapy-related online discussions have not been studied. The aim of this study was to explore content, participants, and structure of physical therapy-related discussions on Twitter. This study was observational. Over 12 weeks, more than 30,000 physical therapy-related tweets were collected. A random sample of 100 tweets underwent preliminary analysis to determine broad categories including tweet author, tone, and theme. A second random sample of 100 tweets was analyzed to confirm categories. Once categories were established, a new sample of 1000 tweets was randomly selected for analysis and categorization. All study investigators categorized a shared collection of tweets to establish inter- and intrarater agreement. Twitter conversations were visualized using NodeXL. Intrarater and interrater agreement for tweet categorization was 95% and 89%, respectively. The distribution of the intended audience was 35.5% professional, 35.5% broad reach, and 29% public. The gross distribution of tweet tone was 63.1% neutral, 31.4% positive, and 5.6% negative. Twenty-eight percent of tweets were authored by physical therapists/physical therapist assistants and nearly one-half were categorized as "marketing." Tweets tended to be "isolated," not within a conversation, or consist of conversation within "tight crowds." This study was purely observational. Social media content can be highly influenced by temporal events, which limits the generalization of specific findings. Study results indicate that the reach of physical therapy-related tweets may not be as broad as intended by the author. Physical therapy professionals and the hospitals/clinics that employ them may need to implement strategies to be more intentional in reaching a broader audience with online messages.

What Does the Public Think? Examining Plastic Surgery Perceptions through the Twitterverse.

Twitter is a recognized social media platform for communication of health information. Rime reported that emotion is the main motive for social sharing. This study is a content analysis of Twitter that was performed to identify the public's perceptions and attitudes toward plastic surgery and the emotional triggers that drive social sharing of plastic surgery information. Tweets containing "#PlasticSurgery" or "Plastic Surgery" were archived randomly from August 1, 2014, to December 30, 2016 (n = 4548). Tweets were categorized according to tweet author, specialty, topic, content, multimedia included, emotion, tone, accuracy of information, source, and retweet rate. Statistical analysis was performed to detect significant patterns. Tweets on cosmetic surgery (74 percent) were shared mostly on Twitter, predominantly posted by the public [n = 1611 (48 percent)]. More than 13 percent of posts contained "celebrity news" and 42.8 percent contained professional information and resources. The most frequent emotions shared and retweeted were "relaxed/content" (51.5 percent) and "excited/interested" (18.4 percent). Most tweets posted by the public contained inaccurate information [n = 1486 (80 percent)]. Only 154 (11.2 percent) of board-certified plastic surgeons' tweets were rated as "most accurate." The majority of tweets posted on Twitter contained inaccurate information that can lead to misperception among the public. Understanding emotional triggers for social sharing provides insight into what is most appealing. To enhance public uptake and sharing of tweets, plastic surgeons can use these findings to promote the specialty using relaxed/content emotions or excitement in their social media posts.

Context-augmented convolutional neural networks for twitter sarcasm detection

Abstract Sarcasm detection on twitter has received increasing research in recent years. However, existing work has two limitations. First, existing work mainly uses discrete models, requiring a large number of manual features, which can be expensive to obtain. Second, most existing work focuses on feature engineering according to the tweet itself, and does not utilize contextual information regarding the target tweet. However, contextual information (e.g. a conversation or the history tweets of the target tweet author) may be available for the target tweet. To address the above two issues, we explore the neural network models for twitter sarcasm detection. Based on convolutional neural network, we propose two different context-augmented neural models for this task. Results on the dataset show that neural models can achieve better performance compared to state-of-the-art discrete models. Meanwhile, the proposed context-augmented neural models can effectively decode sarcastic clues from contextual information, and give a relative improvement in the detection performance.

Quantifying the Twitter Influence of Third Party Commercial Entities versus Healthcare Providers in Thirteen Medical Conferences from 2011 - 2013.

IntroductionTwitter channels are increasingly popular at medical conferences. Many groups, including healthcare providers and third party entities (e.g., pharmaceutical or medical device companies) use these channels to communicate with one another. These channels are unregulated and can allow third party commercial entities to exert an equal or greater amount of Twitter influence than healthcare providers. Third parties can use this influence to promote their products or services instead of sharing unbiased, evidence-based information. In this investigation we quantified the Twitter influence that third party commercial entities had in 13 major medical conferences.MethodsWe analyzed tweets contained in the official Twitter hashtags of thirteen medical conferences from 2011 to 2013. We placed tweet authors into one of four categories based on their account profile: healthcare provider, third party commercial entity, none of the above and unknown. We measured Twitter activity by the number of tweet authors per category and the tweet-to-author ratio by category. We measured Twitter influence by the PageRank of tweet authors by category.ResultsWe analyzed 51159 tweets authored by 8778 Twitter account holders in 13 conferences that were sponsored by 5 medical societies. A quarter of all authors identified themselves as healthcare providers, while only 18% could be identified as third party commercial entities. Healthcare providers had a greater tweet-to-author ratio than their third party commercial entity counterparts (8.98 versus 6.93 tweets). Despite having less authors and composing less tweets, third party commercial entities had a statistically similar PageRank as healthcare providers (0.761 versus 0.797).ConclusionThe Twitter influence of third party commercial entities (PageRank) is similar to that of healthcare providers. This finding is interesting because the number of tweets and third party commercial entity authors required to achieve this PageRank is far fewer than that needed by healthcare providers. Without safety mechanisms in place, the Twitter channels of medical conferences can devolve into a venue for the spread of biased information rather than evidence-based medical knowledge that is expected at live conferences. Continuing to measure the Twitter influence that third parties exert can help conference organizers develop reasonable guidelines for Twitter channel activity.

Portrayal of waterpipe (shisha, hookah, nargile) smoking on Twitter: a qualitative exploration

ObjectivesTo describe and characterize social media content in relation to waterpipe smoking using qualitative methods. Study designExploratory qualitative design. MethodsA representative sample of pre-existing social media content from Twitter relating to waterpipe smoking and written in the English language was collected during a 1 week period in July 2014. A total of 9671 tweets were collected; duplicates and retweets were removed leaving 4439 unique tweets. Data were analyzed semiotically (positive, negative, positive and negative, no sentiment, unclassifiable) and thematically. Photographs attached to tweets written by individual users indexed using #hookah (n = 299) were subjected to content analysis. ResultsOver half of all tweets were positive about waterpipe smoking (59%), with 3% negative, 21% lacking sentiment and 17% unclassifiable. However, there were variations by likely author of tweet, with 91% of tweets from individual users classified as positive. Twitter users focused on their emotional experience, location, other products they were consuming alongside waterpipe smoking, and who they were with. Analysis of photographs highlighted a high degree of synergy between text and visual representations of waterpipe smoking, and two thirds of photographs contained at least part of a waterpipe. ConclusionsWaterpipe smoking may be normalized as an enjoyable activity in this online environment, posing a challenge for public health.

Investigating the Use of Machine Learning Algorithms in Detecting Gender of the Arabic Tweet Author

Twitter is one of the most popular social network sites on the Internet to share opinions and knowledge extensively. Many advertisers use these Tweets to collect some features and attributes of Tweeters to target specific groups of highly engaged people. Gender detection is a sub-field of sentiment analysis for extracting and predicting the gender of a Tweet author. In this paper, we aim to investigate the gender of Tweet authors using different classification mining techniques on Arabic language, such as Naive Bayes (NB), Support vector machine (SVM), Naive Bayes Multinomial (NBM), J48 decision tree, KNN. The results show that the NBM, SVM, and J48 classifiers can achieve accuracy above to 98%, by adding names of Tweet author as a feature. The results also show that the preprocessing approach has negative effect on the accuracy of gender detection. In nutshell, this study shows that the ability of using machine learning classifiers in detecting the gender of Arabic Tweet author.

Use of Twitter at a major national pharmacy conference.

The results of a study of Twitter use by attendees of the 2013 ASHP Midyear Clinical Meeting (MCM) and other interested parties are presented. All messages posted on the social media platform Twitter under the official MCM "hashtag" (#ashpmidyear) during the five-day conference were archived and evaluated. Demographic data on authors of MCM-related tweets were collected by evaluating information provided by Twitter users in their public profiles. The archived messages were classified by content type. A total of 1539 messages originating from 400 unique U.S.- and foreign-based Twitter accounts were posted under the MCM hashtag, an average of 3.84 tweets per account. The estimated rate of conference-related Twitter use by MCM attendees was 1.7%. One third of Twitter users posting conference-related tweets were identified as pharmacists; 86 (21.5%) and 65 (16.25%) tweets originated from accounts held by pharmacy students and pharmaceutical industry representatives, respectively. The authors of MCM-related tweets represented a broad cross-section of pharmacy practice settings and specialties. About 39% of the evaluated Twitter postings were classified as social, with about 31% of postings pertaining to specific MCM educational sessions and nearly 25% classified as advertising. The majority of MCM-related tweets by onsite and remote Twitter users were social in nature or pertained to educational sessions held over the course of the conference.

Ethical research standards in a world of big data

In 2009 Ginsberg et al. reported using Google search query volume to estimate influenza activity in advance of traditional methodologies. It was a groundbreaking example of digital disease detection, and it still remains illustrative of the power of gathering data from the internet for important research. In recent years, the methodologies have been extended to include new topics and data sources; Twitter in particular has been used for surveillance of influenza-like-illnesses, political sentiments, and even behavioral risk factors like sentiments about childhood vaccination programs. As the research landscape continuously changes, the protection of human subjects in online research needs to keep pace. Here we propose a number of guidelines for ensuring that the work done by digital researchers is supported by ethical-use principles. Our proposed guidelines include: 1) Study designs using Twitter-derived data should be transparent and readily available to the public. 2) The context in which a tweet is sent should be respected by researchers. 3) All data that could be used to identify tweet authors, including geolocations, should be secured. 4) No information collected from Twitter should be used to procure more data about tweet authors from other sources. 5) Study designs that require data collection from a few individuals rather than aggregate analysis require Institutional Review Board (IRB) approval. 6) Researchers should adhere to a user’s attempt to control his or her data by respecting privacy settings. As researchers, we believe that a discourse within the research community is needed to ensure protection of research subjects. These guidelines are offered to help start this discourse and to lay the foundations for the ethical use of Twitter data.

Decoding twitter: Surveillance and trends for cardiac arrest and resuscitation communication

Aim of the studyTwitter has over 500 million subscribers but little is known about how it is used to communicate health information. We sought to characterize how Twitter users seek and share information related to cardiac arrest, a time-sensitive cardiovascular condition where initial treatment often relies on public knowledge and response. MethodsTweets published April–May 2011 with keywords cardiac arrest, CPR, AED, resuscitation, heart arrest, sudden death and defib were identified. Tweets were characterized by content, dissemination, and temporal trends. Tweet authors were further characterized by: self-identified background, tweet volume, and followers. ResultsOf 62,163 tweets (15,324, 25%) included resuscitation/cardiac arrest-specific information. These tweets referenced specific cardiac arrest events (1130, 7%), CPR performance or AED use (6896, 44%), resuscitation-related education, research, or news media (7449, 48%), or specific questions about cardiac arrest/resuscitation (270, 2%). Regarding dissemination (1980, 13%) of messages were retweeted. Resuscitation specific tweets primarily occurred on weekdays. Most users (10,282, 93%) contributed three or fewer tweets during the study time frame. Users with more than 15 resuscitation-specific tweets in the study time frame had a mean 1787 followers and most self-identified as having a healthcare affiliation. ConclusionDespite a large volume of tweets, Twitter can be filtered to identify public knowledge and information seeking and sharing about cardiac arrest. To better engage via social media, healthcare providers can distil tweets by user, content, temporal trends, and message dissemination. Further understanding of information shared by the public in this forum could suggest new approaches for improving resuscitation related education.

Twitter‐based information distribution during the 2009 Red River Valley flood threat

Twitter‐based information distribution during the 2009 Red River Valley flood threat