Showing papers by "Nello Cristianini published in 2017"

Content analysis of 150 years of British periodicals

Abstract: Previous studies have shown that it is possible to detect macroscopic patterns of cultural change over periods of centuries by analyzing large textual time series, specifically digitized books. This method promises to empower scholars with a quantitative and data-driven tool to study culture and society, but its power has been limited by the use of data from books and simple analytics based essentially on word counts. This study addresses these problems by assembling a vast corpus of regional newspapers from the United Kingdom, incorporating very fine-grained geographical and temporal information that is not available for books. The corpus spans 150 years and is formed by millions of articles, representing 14% of all British regional outlets of the period. Simple content analysis of this corpus allowed us to detect specific events, like wars, epidemics, coronations, or conclaves, with high accuracy, whereas the use of more refined techniques from artificial intelligence enabled us to move beyond counting words by detecting references to named entities. These techniques allowed us to observe both a systematic underrepresentation and a steady increase of women in the news during the 20th century and the change of geographic focus for various concepts. We also estimate the dates when electricity overtook steam and trains overtook horses as a means of transportation, both around the year 1900, along with observing other cultural transitions. We believe that these data-driven approaches can complement the traditional method of close reading in detecting trends of continuity and change in historical corpora.

Circadian mood variations in Twitter content.

Abstract: Background:Circadian regulation of sleep, cognition, and metabolic state is driven by a central clock, which is in turn entrained by environmental signals. Understanding the circadian regulation of mood, which is vital for coping with day-to-day needs, requires large datasets and has classically utilised subjective reporting.Methods:In this study, we use a massive dataset of over 800 million Twitter messages collected over 4 years in the United Kingdom. We extract robust signals of the changes that happened during the course of the day in the collective expression of emotions and fatigue. We use methods of statistical analysis and Fourier analysis to identify periodic structures, extrema, change-points, and compare the stability of these events across seasons and weekends.Results:We reveal strong, but different, circadian patterns for positive and negative moods. The cycles of fatigue and anger appear remarkably stable across seasons and weekend/weekday boundaries. Positive mood and sadness interact more ...

Seasonal Variation in Collective Mood via Twitter Content and Medical Purchases

Abstract: The analysis of sentiment contained in vast amounts of Twitter messages has reliably shown seasonal patterns of variation in multiple studies, a finding that can have great importance in the understanding of seasonal affective disorders, particularly if related with known seasonal variations in certain hormones. An important question, however, is that of directly linking the signals coming from Twitter with other sources of evidence about average mood changes. Specifically we compare Twitter signals relative to anxiety, sadness, anger, and fatigue with purchase of items related to anxiety, stress and fatigue at a major UK Health and Beauty retailer. Results show that all of these signals are highly correlated and strongly seasonal, being under-expressed in the summer and over-expressed in the other seasons, with interesting differences and similarities across them. Anxiety signals, extracted from both Twitter and from Health product purchases, peak in spring and autumn, and correlate also with the purchase of stress remedies, while Twitter sadness has a peak in the Winter, along with Twitter anger and remedies for fatigue. Surprisingly, purchase of remedies for fatigue do not match the Twitter fatigue, suggesting that perhaps the names we give to these indicators are only approximate indications of what they actually measure. This study contributes both to the clarification of the mood signals contained in social media, and more generally to our understanding of seasonal cycles in collective mood.

The Actors of History: Narrative Network Analysis Reveals the Institutions of Power in British Society Between 1800-1950

Abstract: In this study we analyze a corpus of 35.9 million articles from local British newspapers published between 1800 and 1950, investigating the changing role played by key actors in public life. This involves the role of institutions (such as the Church or Parliament) and individual actors (such as the Monarch). The analysis is performed by transforming the corpus into a narrative network, whose nodes are actors, whose links are actions, and whose communities represent tightly interacting parts of society. We observe how the relative importance of these communities evolves over time, as well as the centrality of various actors. All this provides an automated way to analyze how different actors and institutions shaped public discourse over a time span of 150 years. We discover the role of the Church, Monarchy, Local Government, and the peculiarities of the separation of powers in the United Kingdom. The combination of AI algorithms with tools from the computational social sciences and data-science, is a promising way to address the many open questions of Digital Humanities.

Freudian Slips: Analysing the Internal Representations of a Neural Network from Its Mistakes

Abstract: The use of deep networks has improved the state of the art in various domains of AI, making practical applications possible. At the same time, there are increasing calls to make learning systems more transparent and explainable, due to concerns that they might develop biases in their internal representations that might lead to unintended discrimination, when applied to sensitive personal decisions. The use of vast subsymbolic distributed representations has made this task very difficult. We suggest that we can learn a lot about the biases and the internal representations of a deep network without having to unravel its connections, but by adopting the old psychological approach of analysing its “slips of the tongue”. We demonstrate in a practical example that an analysis of the confusion matrix can reveal that a CNN has represented a biological task in a way that reflects our understanding of taxonomy, inferring more structure than it was requested to by the training algorithm. In particular, we show how a CNN trained to recognise animal families, contains also higher order information about taxa such as the superfamily, parvorder, suborder and order for example. We speculate that various forms of psycho-metric testing for neural networks might provide us insight about their inner workings.