A Novel TDEF1.0 for Making Twitter Accessible for People with Disabilities

Automatic question generation (QG) is a challenging problem in natural language understanding. QG systems are typically built assuming access to a large number of training instances where each instance is a question and its corresponding answer. For a new language, such training instances are hard to obtain making the QG problem even more challenging. Using this as our motivation, we study the reuse of an available large QG dataset in a secondary language (e.g. English) to learn a QG model for a primary language (e.g. Hindi) of interest. For the primary language, we assume access to a large amount of monolingual text but only a small QG dataset. We propose a cross-lingual QG model which uses the following training regime: (i) Unsupervised pretraining of language models in both primary and secondary languages and (ii) joint supervised training for QG in both languages. We demonstrate the efficacy of our proposed approach using two different primary languages, Hindi and Chinese. We also create and release a new question answering dataset for Hindi consisting of 6555 sentences.

Mobile devices like smartphones together with social networks enable people to generate, share, and consume enormous amounts of media content. Common search operations, for example searching for a music clip based on artist name and song title on video platforms such as YouTube, can be achieved both based on potentially shallow human-generated metadata, or based on more profound content analysis, driven by Optical Character Recognition (OCR) or Automatic Speech Recognition (ASR). However, more advanced use cases, such as summaries or compilations of several pieces of media content covering a certain event, are hard, if not impossible to fulfill at large scale. One example of such event can be a keynote speech held at a conference, where, given a stable network connection, media content is published on social networks while the event is still going on. In our thesis, we develop a framework for media content processing, leveraging social networks, utilizing the Web of Data and fine-grained media content addressing schemes like Media Fragments URIs to provide a scalable and sophisticated solution to realize the above use cases: media content summaries and compilations. We evaluate our approach on the entity level against social media platform APIs in conjunction with Linked (Open) Data sources, comparing the current manual approaches against our semi-automated approach. Our proposed framework can be used as an extension for existing video platforms.

For dispute resolution in daily life, tamper-proof data storage and retrieval of log data are important with the incorporation of trustworthy access control for the related users and devices, while giving access to confidential data to the relevant users and maintaining data persistency are two major challenges in information security. This research uses blockchain data structure to maintain data persistency. On the other hand, we propose protocols for the authentication of users (persons and devices) to edge server and edge server to main server. Our proposed framework also provides access to forensic users according to their relevant roles and privilege attributes. For the access control of forensic users, a hybrid attribute and role-based access control (ARBAC) module added with the framework. The proposed framework is composed of an immutable blockchain-based data storage with endpoint authentication and attribute role-based user access control system. We simulate authentication protocols of the framework in AVISPA. Our result analysis shows that several security issues can efficiently be dealt with by the proposed framework.

O presente artigo investiga os papéis do sujeito com relação ao conteúdo midiático nas redes sociais, realizando-se um estudo sobre as trocas realizadas no Twitter através do Tweetdeck. Propõe-se que o indivíduo é um disseminador de mídia, já que a dinâmica das trocas referenciais na web é interpretada pelo ponto de vista do próprio. Suas afinidades que formam seu perfil e suas interações atribuem-lhe identidade, sendo também conteúdo relevante para os contatos que forma em grupos de interesse. São abordados conceitos de cultura, mídia e representação e realiza-se uma pesquisa quantitativa quanto aos hábitos de absorção e processamento de informação no Tweetdeck. Conclui-se que o sujeito pode ter três papéis com relação ao conteúdo midiático: produtor, compartilhador e leitor.

BackgroundAs social media becomes increasingly popular online venues for engaging in communication about public health issues, it is important to understand how users promote knowledge and awareness about specific topics.ObjectiveThe aim of this study is to examine the frequency of discussion and differences by race and ethnicity of cancer-related topics among unique users via Twitter.MethodsTweets were collected from April 1, 2014 through January 21, 2015 using the Twitter public streaming Application Programming Interface (API) to collect 1% of public tweets. Twitter users were classified into racial and ethnic groups using a new text mining approach applied to English-only tweets. Each ethnic group was then analyzed for frequency in cancer-related terms within user timelines, investigated for changes over time and across groups, and measured for statistical significance.ResultsObservable usage patterns of the terms "cancer", "breast cancer", "prostate cancer", and "lung cancer" between Caucasian and African American groups were evident across the study period. We observed some variation in the frequency of term usage during months known to be labeled as cancer awareness months, particularly September, October, and November. Interestingly, we found that of the terms studied, "colorectal cancer" received the least Twitter attention.ConclusionsThe findings of the study provide evidence that social media can serve as a very powerful and important tool in implementing and disseminating critical prevention, screening, and treatment messages to the community in real-time. The study also introduced and tested a new methodology of identifying race and ethnicity among users of the social media. Study findings highlight the potential benefits of social media as a tool in reducing racial and ethnic disparities.

In recent years, social media has become one of the most important political marketing tools. The aim of the research is to determine how university students voting have attitudes towards political messages run across several different social media channels. Undergraduate students in Tekirdağ Central Campus of Namik Kemal University are generating the population of the study. In this research, sample was not taken. Questionnaire form was used as data collecting tool. Data obtained through questionnaire forms were presented as descriptive statistics (frequency, percentage, mean, standard deviation). For differences between group means, T-test and One-way ANOVA were implemented. It was concluded that messages with political content in social media had intensifier effects on present preferences of university student voters and had directive effects on indecisive students. Twitter users had more negative attitudes towards messages with political content in social media.

In recent years, social media has become one of the most important political marketing tools. The aim of the research is to determine how university students voting have attitudes towards political messages run across several different social media channels. Undergraduate students in Tekirdag Central Campus of Namik Kemal University are generating the population of the study. In this research, sample was not taken. Questionnaire form was used as data collecting tool. Data obtained through questionnaire forms were presented as descriptive statistics (frequency, percentage, mean, standard deviation). For differences between group means, T-test and One-way ANOVA were implemented. It was concluded that messages with political content in social media had intensifier effects on present preferences of university student voters and had directive effects on indecisive students. Twitter users had more negative attitudes towards messages with political content in social media.

In this paper, we introduce an agent for recommending information to a user on Twitter, which is one of the most popular microblogging services. For recommending sufficient information for a user, it is important to extract automatically user's interest with accuracy and to collect new information which interests and attracts the user. The agent that is introduced in this paper extracts automatically user's interests from tweets on the timeline of the user and finds the web sites that would provide new information which interests the user from the tweets. The agent selects recommending information from the web sites and posts it on the user's timeline. Experimental results show that our agent is able to recommend sufficient information for users of Twitter in a natural manner.

Tweets summarization aims to find a group of representative tweets for a specific topic. In recent times, there have been several research efforts toward devising a variety of techniques to summarize tweets in Twitter. However, these techniques are either not personal (i.e., consider only tweets in the timeline of a specific user) or are too expensive to be realized on a mobile device. Given that 80% of active Twitter users access the site on mobile devices, in this demonstration we present a lightweight, personalized, on-demand, topic modeling-based tweets summarization engine called TOTEM, designed for such devices. Specifically, TOTEM summarizes most recent tweets on a user's timeline and enables her to visualize and navigate representative topics and associated tweets in a user-friendly tap-and-swipe manner.

Tweets summarization aims to find a group of representative tweets for a specific set of input tweets or a given topic. In recent times, there have been several research efforts toward devising a variety of techniques to summarize tweets in Twitter. However, these techniques are either not personal (that is, consider only tweets in the timeline of a specific user) or are too expensive to be realized on a mobile device. Given that 80% of active Twitter users access the site on mobile devices, in this article we present a lightweight, personal, on‐demand, topic modeling‐based tweets summarization engine called TOTEM, designed for such devices. Specifically, TOTEM first preprocesses recent tweets in a user's timeline and exploits Latent Dirichlet Allocation‐based topic modeling to assign each preprocessed tweet to a topic. Then it generates a ranked list of relevant tweets, a topic label, and a topic summary for each of the topics. Our experimental study with real‐world data sets demonstrates the superiority of TOTEM.

Social media like Twitter have become globally popular in the past decade. Thanks to the high penetration of smartphones, social media users are increasingly going mobile. This trend has contributed to foster various location based services deployed on social media, the success of which heavily depends on the availability and accuracy of users' location information. However, only a very small fraction of tweets in Twitter are geo-tagged. Therefore, it is necessary to infer locations for tweets in order to attain the purpose of those location based services. In this paper, we tackle this problem by scrutinizing Twitter user timelines in a novel fashion. First of all, we split each user's tweet timeline temporally into a number of clusters, each tending to imply a distinct location. Subsequently, we adapt two machine learning models to our setting and design classifiers that classify each tweet cluster into one of the pre-defined location classes at the city level. The Bayes based model focuses on the information gain of words with location implications in the user-generated contents. The convolutional LSTM model treats user-generated contents and their associated locations as sequences and employs bidirectional LSTM and convolution operation to make location inferences. The two models are evaluated on a large set of real Twitter data. The experimental results suggest that our models are effective at inferring locations for non-geotagged tweets and the models outperform the state-of-the-art and alternative approaches significantly in terms of inference accuracy.

This study explores the problem of inferring locations for individual tweets. We scrutinize Twitter user timelines in a novel fashion. First of all, we split each user's tweet timeline temporally into a number of clusters, each tending to imply a distinct location. Subsequently, we adapt machine learning models to our setting and design classifiers that classify each tweet cluster into one of the pre-defined location classes at the city level. Extensive experiments on a large set of real Twitter data suggest that our models are effective at inferring locations for non-geotagged tweets and outperform the state-of-the-art approaches significantly in terms of inference accuracy.

ABSTRACT
 ObjectivesOur group has investigated the occurrence of psychotic(-like) experiences (PLEs) in Twitter posts, namely auditory hallucinations. Tweets classified as potentially related to auditory hallucinations were proportionately higher between 23:00 and 5:00 in comparison to tweets not classified. This may indicate a clinically significant relationship between sleep and PLEs in the general population, a notion supported by the literature. Based on our previous investigation, the current study aimed to explore whether this methodology could be amended to generate datasets regarding sleep experiences in people who self-report a diagnosis of a psychotic disorder.
 ApproachThe current investigation seeks to establish if it is feasible to generate anonymised datasets regarding sleep by extracting information from the timelines of people who self-report a psychotic diagnosis. A text mining method was implemented that utilised rule-based semantic filters that aimed to identify self-reported diagnoses. This focused on occurrences of personal and possessive pronouns to detect the subjectivity of tweets, as well as potential diagnostic verb indicators and any mentions of other related factors. For each diagnostic tweet, we collected information from user timelines. A sleep-related classifier was then implemented, which used lexical features (e.g. bag-of-words, part-of-speech tags) to predict whether a given tweet refers to sleep-related experience.
 ResultsAfter training the classifier on the bag-of-words model, the most informative words which contributed to the performance of the classifier were: ‘sleep’, ‘can’t awake’, ‘never’, ‘stress’. Part-of-speech tags (e.g. verbs, adverbs) were also important features. The classification accuracy of the ‘bag-of-words’ model was better than the ‘part-of-speech’ model.
 Through the method outlined herein, we were able to improve the quality of the generated datasets in comparison to the previous investigation. This methodology also facilitated the mining of individual Twitter users timelines who stated a personal diagnosis. To this end, an additional filter was implemented to identify tweets regarding sleep experience. The potential relationship between sentiment and temporality expressed in diagnosis and sleep experiences are also discussed.
 ConclusionsThe results from this study have implications for mental health research on Twitter. Specifically, the refinements in the methodology enabled retrieval of two high quality datasets regarding psychosis and sleep. Therefore it is feasible other psychosis-related phenomena (e.g. visual hallucinations, delusions, medication) could also be applied as separate filters to create one dataset of psychosis-related experiences within individuals diagnosed with psychosis.

Mainstream media often simply download information posted on social media from unreliable sources without critical thinking and verifying the content. Nowadays, the accuracy of information in all types of media is less important than the number of clicks and views per media content. Due to the fact that there is a proven correlation between the level of media literacy and democratic development of a society, as well as the fact that the media have a great influence on the shaping of public opinion via media framing and agendas, this paper explores the awareness on media literacy of Twitter users in Croatia (primarily journalists) and their perception of their own consumption and dissemination of fake news across social networks and traditional mass media. Although Twitter is not popular in Croatia, it is intensively used by Croatian journalists, and tweets often serve as topics for media processing, which gives Twitter great importance in creating the Croatian media discourse. The research was conducted by encouraging a discussion and undertaking further content analysis of tweets regarding the role of media literacy in combating the impact of fake news, especially on social networks. This was followed by semi-structured in-depth interviews with six journalists who are active Twitter users and content analysis of articles on media literacy and fake news available on Croatian web portals. Triangulation was used in order to reduce the echo chamber effect on the results of the study. The research has shown that respondents, i.e. Twitter users in Croatia – including journalists – are not sufficiently informed about media literacy and its impact on combating fake news. It has also pinpointed their lack of awareness regarding their own consumption and dissemination of fake news across social networks and mass media. The objective of the research was to create a model for analyzing the interrelationship between media literacy and creating fake news from two perspectives: (1) evaluating self-perception of Twitter users via content analysis and in-depth interviews, and (2) analyzing media publications on the topic of media literacy and fake news, whose content reflects the perception of the media on these topics.

Twitter has attracted millions of users that generate a humongous flow of information at constant pace. The research community has thus started proposing tools to extract meaningful information from tweets. In this paper, we take a different angle from the mainstream of previous works: we explicitly target the analysis of the timeline of tweets from single We define a framework - named TUCAN - to compare information offered by the target users over time, and to pinpoint recurrent topics or topics of interest. First, tweets belonging to the same time window are aggregated into songs. Several filtering procedures can be selected to remove stop-words and reduce noise. Then, each pair of bird songs is compared using a similarity score to automatically highlight the most common terms, thus highlighting recurrent or persistent topics. TUCAN can be naturally applied to compare bird song pairs generated from timelines of different users. By showing actual results for both public profiles and anonymous users, we show how TUCAN is useful to highlight meaningful information from a target user's Twitter timeline.