Prior knowledge of human activities from social data

Zhu, Zack; Blanke, Ulf; Calatroni, Alberto; Tröster, Gerhard

doi:10.1145/2493988.2494343

Cited by 6 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhu et al [46] introduced a technique that extracts, filters, and semantically categorizes "tips" from Foursquare service for the area of San Francisco. The text includes a timestamp of the posts and the location information.…”

Section: Detecting Contextmentioning

confidence: 99%

A Netnographic-Based Semantic Analysis of Tweet Contents for Stress Management

Jussila¹,

Alkhammash²,

Alghamdi³

et al. 2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Social media platforms provide new value for markets and research companies. This article explores the use of social media data to enhance customer value propositions. The case study involves a company that develops wearable Internet of Things (IoT) devices and services for stress management. Netnography and semantic annotation for recognizing and categorizing the context of tweets are conducted to gain a better understanding of users' stress management practices. The aim is to analyze the tweets about stress management practices and to identify the context from the tweets. Thereafter, we map the tweets on pleasure and arousal to elicit customer insights. We analyzed a case study of a marketing strategy on the Twitter platform. Participants in the marketing campaign shared photos and texts about their stress management practices. Machine learning techniques were used to evaluate and estimate the emotions and contexts of the tweets posted by the campaign participants. The computational semantic analysis of the tweets was compared to the text analysis of the tweets. The content analysis of only tweet images resulted in 96% accuracy in detecting tweet context, while that of the textual content of tweets yielded an accuracy of 91%. Semantic tagging by Ontotext was able to detect correct tweet context with an accuracy of 50%.

show abstract

Section: Detecting Contextmentioning

confidence: 99%

A Netnographic-Based Semantic Analysis of Tweet Contents for Stress Management

Jussila¹,

Alkhammash²,

Alghamdi³

et al. 2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

show abstract

“…Using the ATUS taxonomy, we structure the infinitely possible number of high-level daily activities into major activity categories. As shown by [29], the inherent bias in social media data is skewed and underrepresents activity categories such as "Caring for household members" or "Religious and spiritual activities". We find a similar pattern in our dataset and adapt the ATUS taxonomy to include the same activity categories as in [ a null class in case an activity is not clearly apparent from the instance information.…”

Section: Labeling For Self-reported Activitiesmentioning

confidence: 99%

Human activity recognition using social media data

Zhu

Blanke

Calatroni

et al. 2013

Proceedings of the 12th International Conference on Mobile and Ubiquitous Multimedia

Self Cite

View full text Add to dashboard Cite

Human activity recognition is a core component of context-aware, ubiquitous computing systems. Traditionally, this task is accomplished by analyzing signals of wearable motion sensors. While such signals can effectively distinguish various low-level activities (e.g. walking or standing), two issues exist: First, high-level activities (e.g. watching movies or attending lectures) are difficult to distinguish from motion data alone. Second, instrumentation of complex body sensor network at population scale is impractical. In this work, we take an alternative approach of leveraging rich, dynamic, and crowd-generated self-report data as the basis for in-situ activity recognition. By treating the user as the "sensor", we make use of implicit signals emitted from natural use of mobile smartphones. Applying an L1-regularized Linear SVM on features derived from textual content, semantic location, and time, we are able to infer 10 meaningful classes of daily life activities with a mean accuracy of up to 83.9%. Our work illustrates a promising first step towards comprehensive, high-level activity recognition using free, crowd-generated, social media data.

show abstract

“…Predictions have been used in various scenarios for activity recognition [9,11,21,27,31]. Recently, researchers in [27] predicted a person's going-out behaviour in order to asses if the person is going to leave the home or not.…”

Section: Using Prior Probabilitiesmentioning

confidence: 99%

Using time use with mobile sensor data

Borazio

Laerhoven

2013

Proceedings of the 12th International Conference on Mobile and Ubiquitous Multimedia

View full text Add to dashboard Cite

Having mobile devices that are capable of finding out what activity the user is doing, has been suggested as an attractive way to alleviate interaction with these platforms, and has been identified as a promising instrument in for instance medical monitoring. Although results of preliminary studies are promising, researchers tend to use high sampling rates in order to obtain adequate recognition rates with a variety of sensors. What is not fully examined yet, are ways to integrate into this the information that does not come from sensors, but lies in vast data bases such as time use surveys. We examine using such statistical information combined with mobile acceleration data to determine 11 activities. We show how sensor and time survey information can be merged, and we evaluate our approach on continuous dayand-night activity data from 17 different users over 14 days each, resulting in a data set of 228 days. We conclude with a series of observations, including the types of activities for which the use of statistical data has particular benefits.

show abstract

Prior knowledge of human activities from social data

Cited by 6 publications

References 4 publications

A Netnographic-Based Semantic Analysis of Tweet Contents for Stress Management

A Netnographic-Based Semantic Analysis of Tweet Contents for Stress Management

Human activity recognition using social media data

Using time use with mobile sensor data

Contact Info

Product

Resources

About