Popular Social Networking Service Research Articles

Comparative Study for Sentiment Analysis of Financial Tweets with Deep Learning Methods

Nowadays, Twitter is one of the most popular social networking services. People post messages called “tweets”, which may contain photos, videos, links and text. With the vast amount of interaction on Twitter, due to its popularity, analyzing Twitter data is of increasing importance. Tweets related to finance can be important indicators for decision makers if analyzed and interpreted in relation to stock market. Financial tweets containing keywords from the BIST100 index were collected and the tweets were tagged as “POSITIVE”, “NEGATIVE” and “NEUTRAL”. Binary and multi-class datasets were created. Word embedding and pre-trained word embedding were used for tweet representation. As classifiers, Neural Network, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU) and GRU-CNN models were used in this study. The best results for binary and multi-class datasets were observed with pre-trained word embedding with the CNN model (83.02%, 72.73%). When word embedding was employed, the Neural Network model had the best results on the multi-class dataset (63.85%) and GRU-CNN had the best results on the binary dataset (80.56%).

DIPA2

The world today is increasingly visual. Many of the most popular online social networking services are largely powered by images, making image privacy protection a critical research topic in the fields of ubiquitous computing, usable security, and human-computer interaction (HCI). One topical issue is understanding privacy-threatening content in images that are shared online. This dataset article introduces DIPA2, an open-sourced image dataset that offers object-level annotations with high-level reasoning properties to show perceptions of privacy among different cultures. DIPA2 provides 5,897 annotations describing perceived privacy risks of 3,347 objects in 1,304 images. The annotations contain the type of the object and four additional privacy metrics: 1) information type indicating what kind of information may leak if the image containing the object is shared, 2) a 7-point Likert item estimating the perceived severity of privacy leakages, and 3) intended recipient scopes when annotators assume they are either image owners or allowing others to repost the image. Our dataset contains unique data from two cultures: We recruited annotators from both Japan and the U.K. to demonstrate the impact of culture on object-level privacy perceptions. In this paper, we first illustrate how we designed and performed the construction of DIPA2, along with data analysis of the collected annotations. Second, we provide two machine-learning baselines to demonstrate how DIPA2 challenges the current image privacy recognition task. DIPA2 facilitates various types of research on image privacy, including machine learning methods inferring privacy threats in complex scenarios, quantitative analysis of cultural influences on privacy preferences, understanding of image sharing behaviors, and promotion of cyber hygiene for general user populations.

Novel cluster set optimization model with unique identifier tagging for twitter data analysis

Blogs, internet forums, social networks, and micro-blogging sites are some of the growing number of places where users can voice their opinions. Opinions on any given product, issue, service, or idea are contained in data, making them a valuable resource in their own right. Popular social networking services like Twitter, Facebook, and Google+ allows expressing views on a variety of topics, participating in discussions, or sending messages to a global user. Twitter sentiment analysis has received a lot of attention recently.Sentiment analysis is finding how a person feels about a topic from their written response about it and it can be separated into positive and negative through its use. Doing so enables to classify the tweets made by a user in to appropriate classification category based on which some decisions can be made. The literature proposed approaches to develop the classifiers on the Twitter datasets. Operations, including tokenization, stop-word removal, and stemming will be performed. NLP converts the text to a machine-readable representation. Artificial Intelligence (AI) combines NLP data to evaluate if a situation is positive or negative. The document’s subjectivity can be identified using ML and NLP techniques to categorize them in to positive, neutral, or negative. Performing sentiment analysis in Twitter data can be tedious due to limited size, unstructured nature, misspellings, slang, and abbreviations. For this task, a Tweet Analyzing Model for Cluster Set Optimization with Unique Identifier Tagging (TAM-CSO-UIT) was built using prospects to determine positive or negative sentiment in tweets obtained from Twitter. This approach assigns a +ve/-ve value to each entry in the Tweet database based on probability assignment using n-gram model. To perform this effectively the tweet dataset is considered as a sliding window of length L. The proposed model accurately analyses and classifies the tweets.

The Influence of Background Music on People’s Judgment of others on Social Media

This study explored whether background music on Social Networking Service (SNS) can influence people’s judgments about others’ personalities. We used the most popular SNS app in Korea and manipulated the background music. Forty-four participants were recruited and randomly assigned one of two conditions: music from either classical or rock genres. After exploring the profile, the participants answered a questionnaire asking them to assess the profile’s owners’ personality. It was found that both male and female participants viewed the profile owner’s personality differently depending on the background music. The implication of personality judgment in an SNS context was discussed.

Do men and women differ in the capability of weaving online social networks: A perspective of gender stereotype activation

This study aims to identify the gender difference in online social networks by comparing the condition that gender label is manifest and gender stereotype can be activated easily, with the condition under which gender is concealed and the activation of stereotype is difficult. The data about the follower-followee networks from 15,118 individual users and 5933 institutional users in Sina Weibo, a popular local social networking service, was used to compare the gender difference. The findings of this study tell a significantly different story with the prior gender studies regarding the disparity of social networks between genders. The higher degree centrality and lower closeness centrality of males demonstrates that males are more likely to enjoy a central position in both the immediate neighborhood and the entire virtual community. The higher betweenness centrality of males suggests that people depend more on males to make connections with other people, therefore males are more powerful than females in “controlling” the formation of ties between random pairs of users. However, females can increase the likelihood of forming a social tie and building up the whole network structure. Furthermore, people prefer to make an online social tie with those who have different gender with self.

Digital ethnography of an online professional learning community based on WeChat for Chinese early childhood teachers

Enhancing the professional development of early childhood education (ECE) teachers is an important step toward the goal of improving children's learning experiences. Building professional learning communities (PLCs) has been validated by previous works in the ECE domains. The proliferation of social media has catalyzed the emergence of online community, bringing with them their distinctive value in teachers' professional development. WeChat is the most popular social networking service in China. However, only limited attention has been paid to how ECE teachers engaged in online communities and how PLCs for ECE teachers are shaped online in the Chinese contexts. Thus, this study aims to investigate how ECE teachers engage in a WeChat-based community and shape the community as a PLC. Multi-sourced data were collected including excerpts and screenshots of discussions in the chat group and webinars, fieldnotes, and interviews with 12 participants. Key findings suggest that: (a) leadership is distributed through the practices of volunteer facilitators, (b) teachers' practices are shared through reflecting on professional reading and inquiring, (c) shared visions attracted and activated teachers in the online context, and (d) supportive relationships tend to mitigate the isolation of Chinese ECE teachers. This study addresses key challenges to the further development of PLCs based on SNS. The insights about how Chinese early childhood educators utilize the basic functions of social media can contribute to better practices for professional development in and beyond China.

Global Local Fusion Neural Network for Multimodal Sentiment Analysis

With the popularity of social networking services, people are increasingly inclined to share their opinions and feelings on social networks, leading to the rapid increase in multimodal posts on various platforms. Therefore, multimodal sentiment analysis has become a crucial research field for exploring users’ emotions. The complex and complementary interactions between images and text greatly heighten the difficulty of sentiment analysis. Previous works conducted rough fusion operations and ignored the study for fine fusion features for the sentiment task, which did not obtain sufficient interactive information for sentiment analysis. This paper proposes a global local fusion neural network (GLFN), which comprehensively considers global and local fusion features, aggregating these features to analyze user sentiment. The model first extracts overall fusion features by attention modules as modality-based global features. Then, coarse-to-fine fusion learning is applied to build local fusion features effectively. Specifically, the cross-modal module is used for rough fusion, and fine-grained fusion is applied to capture the interaction information between objects and words. Finally, we integrate all features to achieve a more reliable prediction. Extensive experimental results, comparisons, and visualization of public datasets demonstrate the effectiveness of the proposed model for multimodal sentiment classification.

Feasibility study on collecting patient-reported outcomes from breast cancer patients using the LINE-ePRO system.

Due to the increasing complexity of cancer treatment, ensuring safety and maintaining the quality of life during treatment are important issues. Patient‐reported outcomes (PROs) in oncology are essential for assessing patient symptoms. A feasibility study was undertaken on breast cancer patients by building a PRO data collection system based on LINE, one of the most popular social network service applications in Japan. In this study, one or more predefined PRO questions for each breast cancer patient's clinical situation were sent to the patient’s LINE application daily. The patient selected a predefined answer by tapping the screen, but no free‐text answers were allowed. Seventy‐three patients were enrolled. The median observation period was 435 days (84‐656 days), and the total number of PROs collected was 16,417, with a mean of 224.9 reports per patient. Patients on adjuvant endocrine therapy were notified of 2.5 questions per week, and the median number of responses per week and response rate were 2.387 (1.687‐11.627) and 95.5%, respectively. Analyzing the results by age group, the number of responses from those aged 60 and above was equal to or higher than that of the younger age group. It was also possible to track each patient’s PROs accurately. These results suggested that the design of the system, based on an application used daily, instead of using specifically prepared applications for collecting electronic PROs, was the reason for the favorable acceptance from patients and the satisfactory response rate from all age groups, including the elderly.

Personality traits and mental health of social networking service users: A cross-sectional exploratory study among Japanese undergraduates

In this study, we investigated personality and mental health variables for users of three popular social networking services (SNS): Twitter, Instagram, and Facebook, among Japanese undergraduate students (N = 1030). Compared to non-users, Twitter users scored significantly lower in extraversion, conscientiousness, openness, and self-esteem, and higher in neuroticism, behavioral inhibition system (BIS) sensitivity, and social comparison orientation. Instagram users scored significantly higher in extraversion, behavioral approach system (BAS) -drive, BAS-reward responsiveness, BAS-fun seeking, narcissism, self-esteem, and subjective happiness, and lower in neuroticism, anxiety, and to a marginal extent, depression. Facebook users scored significantly higher in extraversion, openness, BAS-fun seeking, self-esteem, and to a marginal extent, narcissism. Within SNS users, extraversion, agreeableness, and self-esteem were negatively associated with negative mental health variables, and positively associated with positive mental health variables. Neuroticism was positively associated with negative mental health variables, and negatively associated with positive mental health variables. Further, some associations were found within specific SNS users. Differences in personality and mental health were discussed based on the characteristics of each SNS platform.

Human body structure extraction of anime characters based on deep learning

With the popularity of social networking services and smartphones, photo processing applications have become widely used, and there is a growing interest in more advanced photo processing techniques. In the field of image generation, methods based on adversarial generative networks (GANs) have shown particularly good results. We discussed the anime characters’ body structure extraction methods based on generative adversarial networks (GANs). This research is expected to be applied in the field of drawing that can be used as a tool to significantly reduce learning period or provide a new study method for drawing beginners.

Rejected or ignored?: the effect of social exclusion on Instagram use motivation and behaviour

ABSTRACT Instagram is one of the most popular social networking services worldwide. This exploratory study was conducted to examine the relationships between social exclusion, Instagram usage motivation (self-promotion, social interaction), and Instagram usage behaviour (frequency of photo uploads, degree of interest in photos uploaded by others). A total of 359 Korean Instagram users responded to an online survey. As a result of correlation analysis and path analysis, it was found that being rejected correlated most with social interaction motivation and being ignored correlated with self-promotion motivation. Findings also indicated that self-promotion motivation was highly correlated with the frequency of photo uploads, leading to Instagram users uploading photos of people, food, and objects more frequently. Social interaction motivation showed a close relationship with a degree of interest in photos uploaded by others, leading users to view and respond to photos of people, food, and objects posted by other users. This study is the first study to suggest social exclusion as an antecedent factor that influences Instagram usage motivation and usage behaviour.

Spatiotemporal Information Fusion Method of User and Social Media Activity

Social media check-in data contains a lot of user activity information. Understanding the types of activities and behavior of social media users has important research significance for exploring human mobility and behavior patterns. This paper studies the user activity classification method for Sina Weibo (a very popular Chinese social network service, referred to as “Weibo”), which combines image expression and spatiotemporal data classification technology to realize the identification of the activity behavior represented by the microblog check-in data. Firstly, the user activities represented by the Sina Weibo check-in data are divided into six categories according to POI attribute information: “catering”, “life services”, “campus”, “outdoors”, “entertainment” and “travel”; Then, through the Convolutional Neural Network (CNN) and K-Nearest Neighbor (KNN) classification methods, the image scene information and spatiotemporal information in the check-in data are fused to classify the activity behavior of microblog users. The experimental results show that the proposed method can significantly improve the accuracy of microblog user activity type recognition and provide more effective data support for accurately exploring human behavior activities.

Text classification model for methamphetamine-related tweets in Southeast Asia using dual data preprocessing techniques

<span>Methamphetamine addiction is a prominent problem in Southeast Asia. Drug addicts often discuss illegal activities on popular social networking services. These individuals spread messages on social media as a means of both buying and selling drugs online. This paper proposes a model, the “text classification model of methamphetamine tweets in Southeast Asia” (TMTA), to identify whether a tweet from Southeast Asia is related to methamphetamine abuse. The research addresses the weakness of bag of words (BoW) by introducing BoW and Word2Vec feature selection (BWF) techniques. A domain-based feature selection method was performed using the BoW dataset and Word2Vec. The BWF dataset provided a smaller number of features than the BoW and TF–IDF dataset. We experimented with three candidate classifiers: Support vector machine (SVM), decision tree (J48) and naive bayes (NB). We found that the J48 classifier with the BWF dataset provided the best performance for the TMTA in terms of accuracy (0.815), F-measure (0.818), Kappa (0.528), Matthews correlation coefficient (0.529) and high area under the ROC Curve (0.763). Moreover, TMTA provided the lowest runtime (3.480 seconds) using the J48 with the BWF dataset.</span>

Student Engagement in the Facebook Learning Environment: A Person-Centred Study

Facebook has been used not only as a popular social network service among college students but also as a platform for promoting learning and teaching effectiveness in different subject areas. While previous studies have demonstrated the utility of Facebook for enhancing student engagement, little is known about whether there are different profiles of students with different engagement patterns in the Facebook learning environment. Adopting a person-centred approach, we aim to fill this gap in the current study by identifying unobserved sub-populations of students with respect to their engagement patterns in the Facebook as a supplemental learning platform. Latent profile analysis revealed three engagement profiles, a minimally engaged, a moderately engaged, and a highly engaged profile. Results also suggested that academic disciplines and teacher involvement in the Facebook learning activities were significant predictors of student membership in the three engagement profiles. Our findings offer implications for the design and delivery of Facebook learning activities that cater to different groups of learners with different learning needs.

Understanding Characteristics of Popular Streamers on Live Streaming Platforms: Evidence from Twitch.tv

Live streaming platforms such as Twitch and Periscope have become some of the most popular synchronous social networking services. To attract viewers, streamers are motivated to broadcast exciting video content while actively interacting with viewers. The emerging stream of research on the live streaming community has examined streamers’ motivations and how the viewers react to streamers. However, few studies have focused on understanding the characteristics of popular streamers. Popular streamers create tremendous business value for social media influencer investors, as they have high potential to create persuasive advertisements and endorsements for firms by promoting their products and services. We aim at examining the key characteristics associated with streamers’ viewer base, namely their personality, professionalism, and streaming affordance. Based on text mining and analyses of video content, our results show: (1) certain personality traits (such as openness) are negatively associated with both cumulative and current popularity, (2) professional players are more likely to attract a larger viewer base, and (3) social affordance, including profile building affordance, social connectivity, and social interactivity, is positively associated with both cumulative and current popularity. Our results provide useful insights into measuring and evaluating streamers’ popularity, which can, in turn, generate actionable strategies for social media influencer investors and platform operators.

Design and Implementation of Automated Steganography Image-Detection System for the KakaoTalk Instant Messenger

As the popularity of social network service (SNS) messengers (such as Telegram, WeChat or KakaoTalk) grows rapidly, cyberattackers and cybercriminals start targeting them, and from various media, we can see numerous cyber incidents that have occurred in the SNS messenger platforms. Especially, according to existing studies, a novel type of botnet, which is the so-called steganography-based botnet (stego-botnet), can be constructed and implemented in SNS chat messengers. In the stego-botnet, by using various steganography techniques, every botnet communication and control (C&C) messages are secretly embedded into multimedia files (such as image or video files) frequently shared in the SNS messenger. As a result, the stego-botnet can hide its malicious messages between a bot master and bots much better than existing botnets by avoiding traditional botnet-detection methods without steganography-detection functions. Meanwhile, existing studies have focused on devising and improving steganography-detection algorithms but no studies conducted automated steganography image-detection system although there are a large amount of SNS chatrooms on the Internet and thus may exist many potential steganography images on those chatrooms which need to be inspected for security. Consequently, in this paper, we propose an automated system that detects steganography image files by collecting and inspecting all image files shared in an SNS chatroom based on open image steganography tools. In addition, we implement our proposed system based on two open steganography tools (Stegano and Cryptosteganography) in the KakaoTalk SNS messenger and show our experimental results that validate our proposed automated detection system work successfully according to our design purposes.

SPATIAL PECULIARITIES OF FACEBOOK AND INSTAGRAM SOCIAL NETWORKING SERVICES DISTRIBUTION IN CHERNIVETSKA OBLAST

Problem Statement and Purpose. Social networking services (SNS) are a modern phenomenon in the field of human relations. As differentiated from western science, the issue of SNS spatial distribution is not a matter of high concern in Ukrainian researchers because of a number of objective and subjective reasons. It makes rele­vant this research. The purpose of this research is to identify the features of spatial differentiation of social networking services in Chernivetska oblast. The main task of the work is to analyze the spatial distribution of the number of users and the level of penetration of social networking services Facebook and Instagram by administrative districts and cities of oblast subordination of Chernivetska region.Data & Methods. There are some difficulties with the collection of primary statistical information regarding users of social networking services. To determine the number of Facebook and Instagram users, we used the targeting function – an advertising mechanism that allows to exclude only a portion of target audience that matches same criteria. The criterion we chose was the coverage of the territory. Primary statistics data of the number of social networking service users on the research territory were collected in the first half of February 2020.Results. Facebook is the most popular SNS in Chernivetska oblast. There is a trend to­wards concentration of users' accounts in big cities. 81.6% of accounts are concentrat­ed in Chernivtsі City which is also the regional center. Vyzhnytskyi, Kelmenetskyi, Khotynskyi districts are distinguished among the districts. Facebook is the least used in Putylskyi, Hertsaivskyi districts. The penetration rate of the social networking ser­vice Facebook in Chernivetska oblast is 32.9%. The highest level is in cities of region subordination Chernivtsі, Novodnistrovsk. High levels of penetration are achieved due to the high development of technical infrastructure, through the companies' and shops' accounts, as well as accounts of residents from surrounding villages. In ad­ministrative districts, the penetration rate is much lower. A relatively high percentage of users are in Vyzhnytskyi, Kelmenetskyi, Sokyrianskyi, Khotynskyi districts. The lowest level of penetration of the SNS is in Putylskyi, Hertsaivskyi districts. The second most popular social networking service in Chernivetska oblast is Instagram (28.6% penetration rate). In terms of development, it is inferior to Facebook, and its main users are mostly young people. In the spatial distribution of Instagram users, we can trace the same trends as the Facebook. In general, the distribution of social net­working services such as Facebook and Instagram in Chernivetska oblast corresponds to a hierarchical model of spatial diffusion. There is a trend towards concentration of users in big cities.

A Hierarchical Attention Model for Social Contextual Image Recommendation

Image based social networks are among the most popular social networking services in recent years. With tremendous images uploaded everyday, understanding users' preferences on user-generated images and making recommendations have become an urgent need. In fact, many hybrid models have been proposed to fuse various kinds of side information~(e.g., image visual representation, social network) and user-item historical behavior for enhancing recommendation performance. However, due to the unique characteristics of the user generated images in social image platforms, the previous studies failed to capture the complex aspects that influence users' preferences in a unified framework. Moreover, most of these hybrid models relied on predefined weights in combining different kinds of information, which usually resulted in sub-optimal recommendation performance. To this end, in this paper, we develop a hierarchical attention model for social contextual image recommendation. In addition to basic latent user interest modeling in the popular matrix factorization based recommendation, we identify three key aspects (i.e., upload history, social influence, and owner admiration) that affect each user's latent preferences, where each aspect summarizes a contextual factor from the complex relationships between users and images. After that, we design a hierarchical attention network that naturally mirrors the hierarchical relationship (elements in each aspects level, and the aspect level) of users' latent interests with the identified key aspects. Specifically, by taking embeddings from state-of-the-art deep learning models that are tailored for each kind of data, the hierarchical attention network could learn to attend differently to more or less content. Finally, extensive experimental results on real-world datasets clearly show the superiority of our proposed model.

Integrated Question-Answering System for Natural Disaster Domains Based on Social Media Messages Posted at the Time of Disaster

Natural disasters are events that humans cannot control, and Japan has suffered from many such disasters over its long history. Many of these have caused severe damage to human lives and property. These days, numerous Japanese people have gained considerable experience preparing for disasters and are now striving to predict the effects of disasters using social network services (SNSs) to exchange information in real time. Currently, Twitter is the most popular and powerful SNS tool used for disaster response in Japan because it allows users to exchange and disseminate information quickly. However, since almost all of the Japanese-related content is also written in the Japanese language, which restricts most of its benefits to Japanese people, we feel that it is necessary to create a disaster response system that would help people who do not understand Japanese. Accordingly, this paper presents the framework of a question-answering (QA) system that was developed using a Twitter dataset containing more than nine million tweets compiled during the Osaka North Earthquake that occurred on 18 June 2018. We also studied the structure of the questions posed and developed methods for classifying them into particular categories in order to find answers from the dataset using an ontology, word similarity, keyword frequency, and natural language processing. The experimental results presented herein confirm the accuracy of the answer results generated from our proposed system.

Mental health of family, friends, and co-workers of COVID-19 patients in Japan

As the coronavirus disease 2019 (COVID-19) pandemic rages on, the mental health of both the infected and non-infected is a rising concern. We used administrative survey data (16402 responses in the last two weeks) using a chatbot on LINE, the most popular social networking service (SNS) in Japan, to show that people with COVID-19 patients in a close setting had higher psychological distress level than those without. We believe that the results indicate an urgent need to prioritize the establishment and implementation of mental health and psychosocial support tailored to family, close relatives, and friends of COVID-19 patients.