Section 4: see other Slide Show Christoph F. Eick UH-DAIS Section 4: see other Slide Show Tweet Emotion Mapping: Understanding US Emotions in Time and Space Related to: http://worldhappiness.report/ed/2018/ & http://hedonometer.org Inspired by: https://www.ted.com/talks/hans_rosling_asia_s_rise_how_and_when Great Dismal Swamp, Virginia Data Analysis and Intelligent Systems Lab
Investigated Research Topics Spatio-Temporal Data Analysis Frameworks that Operate on Continuous Functions, such as Density Functions and Interpolation Functions. Spatial and Spatio-temporal Clustering and Hotspot Discovery Algorithms Emotion and Emotion Change Mapping Design of Animations that Summarize the Change of Emotions for a Region of Interest over a Period of Time St^2: Tools for Spatio-temporal Storytelling Also: Create a Website that makes our emotion analysis, animation and data storytelling services available to the public. Not so much: We mostly rely on tools, designed by other, for the assessment of emotions expressed in text document; e.g. tweets; that is, this subject is not the main focus of our research.
Project Goals Given a set of tweets with the location (longitude and latitude), time they were posted and their emotional assessment in [-1,+1] (+1:=very positive emotions, 0:=no emotions, -1: very negative emotions) Research Goals: Identify spatial regions of highly positive emotions (e.g. average emotional assessment >0.4) and regions of highly negative emotions (e.g. average emotion assessment < -0.4) for an observation period; e.g. a year. Identify spatial regions of high discrepancies in emotions (variance of the emotional assessment values is high for the tweets in the region) for an observation period. Identify dominant topics in the tweets identified in steps 1 and 2. Capture temporal patterns of the evolution of the regions identified in step 1 and 2 including: Continuity over time Disappearance / Appearance of new regions Growth / Shrinkage Seasonality Drastic Variations in Emotions in Short Periods of Time … Convert results found in steps 1, 2, 4 into movies/animations that capture the spatio-temporal evolution of emotions in an observation region (e.g. Texas, US,…) over a period of time (e.g. 5 years, 1 year, 1 month): kind of facilitating (spatio-temporal) data storytelling; similar to: https://www.ted.com/talks/hans_rosling_asia_s_rise_how_and_when
Tweet Emotion Analysis http://users.humboldt.edu/mstephens/hate/hate_map.html
Tweet Analysis
http://www.fearofcrime.com/
https://www.tableau.com/solutions/customer/storytelling-data-0 https://www.encorebusiness.com/blog/tableau-tips-tricks-tableau-story-telling/ Comment: Tableau seems to be completely crazy about using their tool about Data Storytelling. However, at the moment I do not see any convincing evidence that their tools do anything sensational with respect to data storytelling or with respect to data analytics: unless they present something more convincing evidence people will continue to use R tools, ArcGIS maybe Python tools for Data Science and not Tableau.
Display Tools1 for St^2: Bubble Charts Gap Minder: https://www.gapminder.org/tools/#_data_/_lastModified:1521732308820;&chart-type=bubbles Video: https://www.ted.com/talks/hans_rosling_asia_s_rise_how_and_when Bubble Chart r Data Analysis and Intelligent Systems Lab
Display Tools1 for St^2: Bubble Charts Display Tools for St^2: Density Polygon Trees proposed by Y. Zhang & Ch. Eick at ACM SIGSPATIAL 2017 Conference r Data Analysis and Intelligent Systems Lab
Display Tools1 for St^2: Bubble Charts E-R Diagram for Change Analysis Output from to # name longitude latitude time ems text C-Predicate Contains Tweet Batch Belongs-to BC-relation-ship C-Relation-ship C-Predicate name from CC-relation-ship Spatial Cluster r to polygon ems--avg den-val Data Analysis and Intelligent Systems Lab
http://hedonometer.org/index.html
Tweet-Emo-Meter Research Related: http://worldhappiness.report/ed/2018/