FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2340743819> ?p ?o ?g. }

• W2340743819 abstract "Design Problem Acknowledgements Testing Finding Specific, Topic Related Information from a Sea of Social Media Posts Scott Duberstein, WSU Undergraduate Student (Presenter), Daniel Asamoah, Ph.D. (PI), Derek Doran, Ph.D. (Co-PI), Shu Schiller, Ph.D. (Co-PI) Given the high traffic volume of Twitter and the infinite access of timeline information one has with the Twitter API, identifying bots and accounts attempting to sell and promote products, from humans who have had genuine experiences with is crucial. Determining who had real information became a key sticking point. Content accounts, accounts that tweet only on a specific topic, are accounts that do both evaluation and gathering of supported medical advice without searching with the Twitter API. They are actively retweeting, tweeting advice, and sharing links with “useful” information. While the Twitter API does a thorough job of returning queries, outside influences such as the current political climate inject seemingly relevant, yet extraneous and weightless components to the data collection process culture increasingly talk about in such ways that are to collect and perform post processing on. Ultimately determining who was is telling a story and sharing information with regards to their audience and what was their motivation brings up a lot of challenges that are not just technical, but human as well. Understanding how people talk about how their medicine is a key feature. using colloquial terms so instead of looking for Duloxetine, gathering tweets that contain the word Cymbalta. So like terms for Dextroamphetamine would be ADD medication or Adderrall or Addy. Users are more likely to be use these terms because of Twitter's character limit restriction as well as the colloquial and societal labels associated with them. Implementation During querying the Twitter API tweets are evaluated to determine whether or not they should be stored in a PostgreSQL database. Setting flags at the start of an initial query is paramount to negating bots. Manual retweets are neglected from any results as these make up for a vast amount of tweets made by bots. Before writing to the database the tweets are passed through a stop word dictionary. The dictionary contains pop culture references in order to choose accounts that are attempting to tell a real story as opposed to ones that are picking on celebrities or politicians who,”forgot to take their meds”. The evaluation process tests if the database has too many instances of Kanye West and Lexapro, then it is safe to assume the tweet is a lyric and remove it safely because it contains no relevant content. This rule applies with accounts that tweet the same material over and over. Pulling a tweet is a random event, if an instance of the tweet occurs with the same text and a new ID, we can skip the user entirely and remove them. There is no clear motivation as to why someone would tweet the same content over and over. If it is a new tweet with new text and a new ID, the database already has a reference to the account, so do not bother writing. We just update the number of unique instances and examine whether or not the account is a content account or person with relevant data. 1. Incorporate the use of crowd sourcing with Amazon Mechanical Turk to determine sentiment and truth to each tweet's validity. 2. Use natural language processing to better make sense of tweets that are pulled from the same hashtag. Where some may feel #depressed about exams others may actually be talking about what it actually feels like to be depressed. 3. Identify social networks that are actively supporting each other and exchanging information through the use of ego networks. 4. Analyze sentiment over time given a tweet that states success or failure with a particular drug or technique with coping using both Amazon Mechanical Turk and algorithms. Fig 2: Google Trends on topics associated misspellings with Adderall Future Goals In order to better understand what hashtags are relevant when collecting and searching with the Twitter API, looking to visualizations such as those provided with RiteTag proved to be most beneficial. Using these services painted a clearer picture as to what topics were trending and had stories with users who were actively talking about their experiences with and mental illness treatments. Fig 1: RiteTag Hashtag Visualization Once implementation was complete and queries were returning positive results, the real human side began to show. Even at its intended core, the “real” content, the stories that mattered are not necessarily taking place on content accounts or via hashtags, but via @ replies. People are having more intimate conversations with one another through @ replies. @ replies make up more than 45 percent of 400 tweets related to just prescription drugs. Fig 3: Example of a Tweet tagged as an @ Reply from the RX Query List As social media continues to become an incredible mode of communication in daily life dealing with the exchange of information, these systems provide authors a platform where they can share their thoughts, feelings, and experiences about a number of topics. Harnessing the information expressed publicly through these modes can be incredibly powerful: public perceptions, signals, and data about a variety of specific topics could be extracted and studied from these posts. However, there is a common trade-off in collecting information about a topic from social media: the more specific the topic, generally, the more challenging it is to extract meaningful information. This is because, at first glance, social media posts are simply too noisy: authors post topics that are forced to inject meaning in a short length (140 characters on Twitter). This work presents a nontrivial methodology to overcome this problem. It uses state-of-the-art programming and data storage technologies, stop-word dictionaries, author filters and Twitter bot detectors. Short of evaluating the authenticity of the collected tweets, which will be done in future work through Amazon Mechanical Turk evaluators, we demonstrate how our methodology extracts specific, meaningful tweets about topics related to chronic diseases and medication. Fig 4: Ego-network of a user (red, pointed) in a social network" @default.
• W2340743819 created "2016-06-24" @default.
• W2340743819 creator A5015828465 @default.
• W2340743819 creator A5024424116 @default.
• W2340743819 creator A5072545173 @default.
• W2340743819 creator A5077752591 @default.
• W2340743819 date "2016-01-01" @default.
• W2340743819 modified "2023-09-27" @default.
• W2340743819 title "Finding Specific , Topic Related Information from a Sea of Social Media Posts" @default.
• W2340743819 hasPublicationYear "2016" @default.
• W2340743819 type Work @default.
• W2340743819 sameAs 2340743819 @default.
• W2340743819 citedByCount "0" @default.
• W2340743819 crossrefType "journal-article" @default.
• W2340743819 hasAuthorship W2340743819A5015828465 @default.
• W2340743819 hasAuthorship W2340743819A5024424116 @default.
• W2340743819 hasAuthorship W2340743819A5072545173 @default.
• W2340743819 hasAuthorship W2340743819A5077752591 @default.
• W2340743819 hasConcept C108827166 @default.
• W2340743819 hasConcept C112698675 @default.
• W2340743819 hasConcept C136764020 @default.
• W2340743819 hasConcept C144133560 @default.
• W2340743819 hasConcept C41008148 @default.
• W2340743819 hasConcept C518677369 @default.
• W2340743819 hasConceptScore W2340743819C108827166 @default.
• W2340743819 hasConceptScore W2340743819C112698675 @default.
• W2340743819 hasConceptScore W2340743819C136764020 @default.
• W2340743819 hasConceptScore W2340743819C144133560 @default.
• W2340743819 hasConceptScore W2340743819C41008148 @default.
• W2340743819 hasConceptScore W2340743819C518677369 @default.
• W2340743819 hasLocation W23407438191 @default.
• W2340743819 hasOpenAccess W2340743819 @default.
• W2340743819 hasPrimaryLocation W23407438191 @default.
• W2340743819 hasRelatedWork W2283717573 @default.
• W2340743819 hasRelatedWork W2529561961 @default.
• W2340743819 hasRelatedWork W2747414241 @default.
• W2340743819 hasRelatedWork W2775232093 @default.
• W2340743819 hasRelatedWork W2900606786 @default.
• W2340743819 hasRelatedWork W2943087902 @default.
• W2340743819 hasRelatedWork W3003937485 @default.
• W2340743819 hasRelatedWork W3008562858 @default.
• W2340743819 hasRelatedWork W3019243142 @default.
• W2340743819 hasRelatedWork W3097245960 @default.
• W2340743819 hasRelatedWork W3106474143 @default.
• W2340743819 hasRelatedWork W3124376805 @default.
• W2340743819 hasRelatedWork W3136339167 @default.
• W2340743819 hasRelatedWork W3160851452 @default.
• W2340743819 hasRelatedWork W3184845153 @default.
• W2340743819 hasRelatedWork W3191669782 @default.
• W2340743819 hasRelatedWork W3195909742 @default.
• W2340743819 hasRelatedWork W3208903131 @default.
• W2340743819 hasRelatedWork W3211384047 @default.
• W2340743819 hasRelatedWork W870878847 @default.
• W2340743819 isParatext "false" @default.
• W2340743819 isRetracted "false" @default.
• W2340743819 magId "2340743819" @default.
• W2340743819 workType "article" @default.