Social Media offer a vast amount of geo-located and time-stamped textual content directly generated by people. This information can be analysed to obtain insights about the general state of a large population of users and to address scientific questions from a diversity of disciplines. In this work, we estimate temporal patterns of mood variation through the use of emotionally loaded words contained in Twitter messages, possibly reflecting underlying circadian and seasonal rhythms in the mood of the users. We present a method for computing mood scores from text using affective word taxonomies, and apply it to millions of tweets collected in the United Kingdom during the seasons of summer and winter. Our analysis results in the detection of strong and statistically significant circadian patterns for all the investigated mood types. Seasonal variation does not seem to register any important divergence in the signals, but a periodic oscillation within a 24-hour period is identified for each mood type. The main common characteristic for all emotions is their mid-morning peak, however their mood score patterns differ in the evenings.
Οι «αριστεροί» αυτή τη στιγμή μπορεί να κερδίζουν την (υπό-) μάχη των Blogs και των Social Media (με μικρή ίσως διαφορά), αλλά οι «νεοφιλελεύθεροι» κερδίζουν τη συνολική μάχη των ΜΜΕ (αφού ορίζουν τα περισσότερα).
Το πρόβλημα είναι ότι πολύ σύντομα οι «νεοφιλελεύθεροι» θα είναι σε θέση να κερδίσουν και την επικοινωνιακή μάχη των Social Media, αν, δηλαδή, δεν τα έχουν ήδη καταφέρει.
Αυτό που θέλω να πω είναι ότι ο σκέτος επικοινωνιακός πόλεμος των «αριστερών» είναι καταδικασμένος σε ολική αποτυχία. Και η απορία μου είναι πώς γίνεται οι «αριστεροί» να μη το βλέπουν κι αυτό.
Title: Detecting Events and Patterns in the Social Web with Statistical Learning
Abstract: A vast amount of textual web streams is influenced by events or phenomena emerging in the real world. The Social Web forms an excellent modern paradigm, where unstructured user generated content is published on a regular basis and in most occasions is freely distributed. The main purpose of this talk is to present methods that enable us to automatically extract useful conclusions from this raw information in both supervised and unsupervised learning scenarios. Our input data stream will be the micro-blogging service of Twitter and presented applications will include the 'nowcasting' of Influenza-like illness rates as well as collective mood analysis for the UK.
(For a more formal report on this, please refer to Section 6.2 of my Ph.D. Thesis.)
2011 was an interesting year for many and obvious reasons. Well, this is my opinion. However, a year is perceived very differently by individuals and therefore, the mean tendencies in the population are of interest.
In this direction ‒ extracting emotional tendencies from the general population ‒, I have analysed Twitter content geolocated in the UK and have extracted affective (mood) scores using an approach which is presented in my Ph.D. Thesis. Four emotions have been considered, namely anger, fear, sadness and joy. In the attached plot, the scores of anger, fear and sadness have been merged and joy's scores have been subtracted from them. Red line is the exact extracted signal, whereas the black line is its smoothed version (using a 7-point moving average to derive a weekly tendency). read more »