Identifying Strain Related On Public Connections In Online Social Networks
ABSTRACT: Psychological stress is threatening people’s health. It is non-trivial to detect stress timely for proactive care. With the popularity of social media, people are used to sharing their daily activities and interacting with friends on social media platforms, making it feasible to leverage online social network data for stress detection. In this paper, we ?nd that users stress state is closely related to that of his/her friends in social media, and we employ a large-scale dataset from real-world social platforms to systematically study the correlation of users’ stress states and social interactions. We ?rst de?ne a set of stress-related textual, visual, and social attributes from various aspects, and then propose a plot. Experimental results show that the proposed model can improve the detection performance. With the help of enumeration we build a website for the users to identify their stress rate level and can check other related activities.
Keywords: Stress Detection, Micro-blog, Social media, Social interaction, Factor graph model.
Psychological Stress is becoming a Threat to People’s Health Nowadays. With the rapid pace of life, more and more people are feeling stressed. Over half of the population have experienced an appreciable rise in stress over the last two years. Though stress itself is non-clinical and common in our life, excessive and chronic stress can be rather harmful to people’s physical and mental health. According to existing research works, long-term stress has been found to be related to many diseases, e.g., clinical depressions, insomnia etc. the development of social networks like Twitter and Facebook, more and more people are willing to share their daily events and moods, and interact with friends through the social networks. As these social media data timely reflect users’ real life states and emotions in a timely manner, it offers new opportunities for representing, measuring, modeling, and mining users behavior patterns through the large-scale social networks, and such social information can find its theoretical basis in psychology research. Limitations Exist in Tweeting Content Based Stress Detection .First, tweets are limited to a maximum of 140 characters on social platforms like Twitter, and users do not always express their stressful states directly in tweets. Second, users with high psychological stress may exhibit low activeness on social networks. These phenomena incur the inherent data sparsity and ambiguity problem, which may hurt the performance of tweeting content based stress detection performance. The tweet contains only 13 characters, saying that the user wished to go home for the Spring Festival holiday. Although no stress is revealed from the tweet itself, from the follow-up interactive comments made by the user and her friends, we can find that the user is actually stressed from work. Thus, simply relying on a user’s tweeting content for stress detection is insufficient. Users’ Social Interactions on Social Networks Contain Useful Cues for Stress Detection. Social psychological studies have made two interesting observations. The first is mood contagion: a bad mood can be transferred from one person to another during social interaction. The second is linguistic echoes: people are known to mimic the style and affect of another person. These observations motivate us to expand the scope of tweet wise investigation by incorporating follow up social interactions like comments and re-tweeting activities in user’s stress detection. This may actually help to mitigate the single user’s data sparsity problem. Another reason for considering social interactions in stress detection is based on our empirical findings on a large-scale dataset. The social structures of stressed users are less connected and thus less complicated than those of non-stressed users. This is consistent with the Pew Research Center’s finding that stressed users are less active than non-stressed ones.
Psychological stress detection is related to the topics of sentiment analysis and emotion detection. Research on tweet-level emotion detection in social networks. Computer-aided detection, analysis, and application of emotion, especially in social networks, have drawn much attention in recent years. Many studies on social media based emotion analysis are at the tweet level, using text-based linguistic features and classic classification approaches. Proposed a system called Mood Lens to perform emotion analysis on the Chinese micro-blog platform Weibo, classifying the emotion categories into four types, i.e., angry, disgusting, joyful, and sad. The emotion propagation problem in social networks, and found that anger has a stronger correlation among different users than joy, indicating that negative emotions could spread more quickly and broadly in the network. As stress is mostly considered as a negative emotion, this conclusion can help us in combining the social influence of users for stress detection. However, these work mainly leverage the textual contents in social networks. In reality, data in social networks is usually composed of sequential and inter-connected items from diverse sources and modalities, making it be actually cross-media data. While tweet-level emotion detection reflects the instant emotion expressed in a single tweet, people’s emotion or psychological stress states are usually more enduring, changing over different time periods. In recent years, extensive research starts to focus on user-level emotion detection in social networks. Our recent work proposed to detect users psychological stress states from social media by learning user-level presentation via a deep convolution network on sequential tweet series in a certain time period. Motivated by the principle of homophily, incorporated social relationships to improve user-level sentiment analysis in Twitter. Though some user-level emotion detection studies have been done, the role that social relationships plays in one’s psychological stress states, and how we can incorporate such information into stress detection have not been examined yet .Research On Leveraging Social Interactions For Social Media Analysis. Social interaction is one of the most important features of social media platforms. Now many researchers are focusing on leveraging social interaction information to help improve the effectiveness of social media analysis. The relationships between social interactions and users’ thinking and behaviors, and found out that Twitter-based interaction can trigger effectual cognitions. However, these work mainly focused on the content of social interactions, e.g., textual comment content, while ignoring the inherent structural information like how users are connected.
Inspired by psychological theories, we first define a set of attributes for stress detection from tweet-level and user-level aspects respectively: 1) tweet-level attributes from content of user’s single tweet, and 2) user-level attributes from user’s weekly tweets. The tweet-level attributes are mainly composed of linguistic, visual, and social attention (i.e., being liked, re-tweeted, or commented) attributes extracted from a single-tweet’s text, image, and attention list. The user-level attributes however are composed of: (a) posting behaviour attributes as summarized from a user’s weekly tweet postings; and (b) social interaction attributes extracted from a user’s social interactions with friends. In particular, the social interaction attributes can further be broken into: (i) social interaction content attributes extracted from the content of users’ social interactions with friends; and (ii) social interaction structure attributes extracted from the structures of users’ social interactions with friends.
Two challenges exist in psychological stress detection. 1) How to extract user-level attributes from user’s tweeting series and deal with the problem of absence of modality in the tweets? 2) How to fully leverage social interaction, including interaction content and structure patterns, for stress detection? To tackle these challenges, we propose a novel hybrid model by combining a factor graph model with a convolutional neural network (CNN), since CNN is capable of learning unified latent features from multiple modalities, and factor graph model is good at modeling the correlations. In this section, we will first introduce the architecture of our model, and then describe the details of each part of the proposed model.
Fig: 1 architectureArchitecture of our mode: The model consists of two parts. The first part is a CNN. The second part is a FGM. The CNN will generate user- level content attributes by convolution with CAE filters as input to the FGM. Take the user labeled with a red star as example. Tweet-level attributes of the user are processed through a convolution with CAE to form the user-level content attributes. The user-level attributes are denoted by xt in the left box. Every xt contains three aspects: user-level content attributes, user-level posting behavior attributes, and user-level social interaction attributes. Data of other users follows the same route. In the FGM, attribute factors connect user-level attributes to corresponding stress states. Social factors connect the stress state of different users. Dynamic factors connect stress state of a user over time. The output of the user’s user-level stress state at time t is yt as highlighted in red, which actually denotes the stress state of the user in weekly period in this paper.
Content of social interaction refers to the content of tweets, comments and re-tweets, including text, emotions, and punctuation marks. We extract emotional words from the interaction content of tweets, and categorize the extracted words into corresponding groups. We compare the frequencies of different word categories between stressed and non-stressed users. The comparison results of the most widely used word categories in our data set, we observe that there is an obvious difference in interaction contents between stressed and non-stressed users. That is, interaction contents of stressed users’ tweets contains much more words from categories like death, sadness, anxiety, anger, and negative emotion, while non-stressed users’ tweets contain more words from categories like friends, family, affection, leisure, and positive emotion.
We first define two sets of attributes to measure the differences of the stressed and non-stressed users on social media platforms:1) tweet-level attributes from a user’s single tweet; 2) user- level attributes summarized from a user’s weekly tweets.
Tweet-level attributes describe the linguistic and visual content, as well as social attention factors (being liked, commented, and re-tweeted) of a single tweet. which categorizes words based on their linguistic or psychological meanings, so we can classify words into different categories, e.g., positive/negative emotion words, degree adverbs. Furthermore, we extract linguistic attributes of emoticons and punctuation marks (‘!’, ‘?’, ‘…’, ‘.’) defines every emoticon in square brackets (e.g., they use ha ha for “laugh”), so we can map the keyword in square brackets to find the emoticons. Twitter adopts Unicode as the representation for all emojis which can be extracted directly.
Compared to tweet-level attributes extracted from a single tweet, user-level attributes are extracted from a list of user’s tweets in a specific sampling period. psychological stress often results from cumulative events or mental states. On the other hand, users may express their chronic stress in a series of tweets rather than one. Besides, the a forementioned social interaction patterns of users in a period of time also contain useful information for stress detection. Moreover, as aforementioned, the information in tweets is limited and sparse, we need to integrate more complementary information around tweets, e.g., users’ social interactions with friends. Thus, appropriately designed user-level attributes can provide a macro-scope of a user’s stress states, and avoid noise or missing data. Here, we define user-level attributes from two aspects to measure the differences between stressed and non- stressed states based on users’ weekly tweet postings: 1) user- level posting behavior attributes from the user’s weekly tweet postings; and 2) user-level social interaction attributes from the user’s social interactions beneath his/her weekly tweet postings.
We presented a framework for detecting users psychological stress states from users’ weekly social media data, leveraging tweets’ content as well as users’ social interactions. Employing real-world social media data as the basis, we studied the correlation between user’ psychological stress states and their social interaction behaviors. In this work, we also discovered several intriguing phenomena of stress.FUTUREWORK
The future scope of the project is to develop a system that not only detecting the stress and also able to analyze people mind means that it will play as a survey system. So that it may provide a better solution on behalf of people of the society for every debatable concepts and also it will indirectly play an important role in political, government and also social media. So we may efficiently analyze stress and also find solution to every social issue by means of polling and analyzing comments.REFERENCES:
1 Golnoosh Faradic, Geetha Sitaraman, Shanu Sushmita, Fabio Celli, Michal Kosinski, David Stillwell, Sergio Davalos, Marie Francine Moens, and Martine De Cock. Computational personality recognition in social media. User Modeling and User- Adapted Interaction, pages 1–34, 2016.
2 Eileen Fischer and A. Rebecca Reuber. Social interaction via new social media: (how) can interactions on twitter affect effectual thinking and behavior? Journal of Business Venturing, 26(1):1–18, 2011.
3 Quan Goo, Jia Jia, Guangyao Shen, Lei Zhang, Lianhong Cai, and Zhang Yi. Learning robust uniform features for cross-media social data by using cross autoencoders. Knowledge Based System, 102:64– 75, 2016.
4 David W. Hosmer, Stanley Lemeshow, and Rodney X. Sturdivant. Applied logistic regression. Wiley series in probability and mathematical statistics, 2013.5 Frank R Kschischang, Brendan J Frey, and H-A Lonelier. Factor graphs and the sum-product algorithm. Information Theory, IEEE Transactions on, 47(2):498–519, 2001.
6 YenLacuna and JoshuaBagnio. Convolutional networks for images, speech, and time series. The handbook of brain theory and neural networks, 3361, 1995.7 Li Liu and Ling Shao. Learning discriminative representations from rgb-d video data. In Proceedings of International Joint Conference on Artificial Intelligence, pages 1493–1500, 2013.8 H-A Loeliger. An introduction to factor graphs. Signal ProcessingMagazine, IEEE, 21(1):28–41, 2004.
9 Federico Alberto Pozzi, Daniele Maccagnola, Elisabetta Ferine, and Enza Messina. Enhance user-level sentiment analysis on microblogs with approval relations. In AI* IA 2013: Advances in Artificial Intelligence, pages 133–144. 2013