9607 research outputs found
Sort by
Prediction
No full textPrediction has a long history in the social sciences, and advances in comput-ing and statistics have transformed our ability to predict in a wide range of domains.However, concerns have been raised about an indiscriminate application of a predic-tive logic, and crime is an area where this is quite pronounced. Indeed, while the po-lice, correctional service, and criminal courts have become increasingly reliant on dig-ital systems of prediction, critics have drawn our attention to numerous issues andcomplexities attendant to this process. This chapter looks at prediction in the crimino-logical realm and provides an overview of key arguments concerning the way data aregenerated, organized, and used as input for predictive tools and technologies, and howthe results are interpreted in the context of criminal justice. By doing so, it aims toshow that the discussions surrounding prediction highlight how digital tools are trans-forming the nature of knowledge and expertise within the criminal justice syste
High-dimensional density-based clustering using locality-sensitive hashing
No full textThe DBSCAN algorithm is a popular density-based clustering method to find clusters of arbitrary shapes without requiring an initial guess on the number of clusters. While there are methods to run DBSCAN efficiently in low-dimensional data in near-linear time, there remains a need for an efficient DBSCAN algorithm that scales to high-dimensional data. The bottleneck in highdimensional data is that the range queries necessary in carrying out the algorithm suffer from the curse of dimensionality. In this paper we present the SRRDBSCAN algorithm. This algorithm is an implementation of approximate DBSCAN using locality-sensitive hashing. We prove sub-quadratic running time bounds under reasonable assumptions about the data. An important ingredient in the design of the data structure is the use of a multi-level LSH data structure, which automatically adapts to the density of data points. An extensive empirical analysis shows that the approximation does not significantly impact the quality of the clustering found by the algorithm as compared to the exact DBSCAN clustering. Moreover, our algorithm is competitive with other approaches even in low-dimensional settings, and thus provides a general-purpose DBSCAN implementation for arbitrary data
The Digital Leviathan: Prediction, Politics and Police Power in POL-INTEL
Get PDFIn this digital era, police forces across the globe are turning to cutting-edge data analytics for the purpose of enacting more efficient police power through predicting and pre-empting crime. In Denmark, allegedly one of the most digitalized countries in the world, the police have turned to the American firm Palantir Technologies to produce a platform named POL-INTEL to integrate, analyze and visualize mass amounts of data from different data bases. For the Danish national police, this platform was heralded as a “super weapon” with predictive policing capacities that would represent a “quantum leap” into the future of law enforcement. For critics, POL-INTEL has been branded a tool of mass surveillance. With this background in mind, this thesis asks two questions: How is police power imagined and enacted in the digital era? And how is governance over the police materialized in relation to data-driven policing? To answer these questions, this thesis develops a methodological framework that combines ethnographic, historiographic and interventionist approaches. Ethnographically, data is drawn from interviews with police officers as well as a variety of other actors, while following the data of those profiled by the police through the criminal justice system. Furthermore, a variety of documents, ranging from public accounts in the press serving to detail the public debate, to internal police handbooks, state reports, etc. are featured. In terms of theory, this thesis synthesizes concepts from critical theory and Science and Technology Studies in particular, alongside Critical Data Studies, police studies with a particular focus on predictive policing, as well as critical criminology. Together, these produce a useful framework for analyzing the complexities of police power, and the materialization of governance, on multiple different levels. Specifically, the thesis investigates the history of police power, tracing how police power has been imagined from the 17th century to the modern notion of predictive policing. POL-INTEL constitutes a case of digital police technology that is expected and portrayed as if it brings immense efficiency in producing social order through the application of science and technology. Through this investigation, the thesis historically ties the notion of predictive policing to the state in a way that has generally been obscured in earlier literature. Concretely, the thesis argues that predictive techniques and technologies have been a major element in the enactment of police power throughout history and follows how the specific notion of predictive policing has been revised and demarcated in the modern era, which has created conceptual inflation. In contrast, the notion of “prediction in action” is launched as a way of capturing the variety of ways law enforcement attempts to predict across different sites and with different technologies. Moreover, the thesis shows that police power has been imagined through predictive data analytics such as POL-INTEL in ways that conflict with how police power is enacted in practice, where the promised effects and new working methods are rarely fully implemented or successful. Instead, the thesis shows that the ways police power is imagined are ideological and serve to black box the enactment of police power. In turn, this black boxing means that police are able to hide their own biases, practices and politics, as well as how they influence the state itself by strategically navigating those forms of governance materialized to control law enforcement. This discovery reverses classical philosophical schemas of police as subordinate to the state and underlines instead how police power may influence government institutions and elected politicians. Details of the complexities, contradictions, and nuances of how police power is enacted in the digital era and how governance over the police is materialized in relation to data-driven policing are also explored. For instance, this thesis described in depth the internal conflicts and contradictions within the police regarding POL-INTEL as a managerial tool that attempts to curtail, limit or direct police discretion. At the same time, the thesis underlines how police discretionary power is still a significant factor in Danish law enforcement with racially biased police profiling practices feeding biased data into the platform. By utilizing and developing the concept of feedback loops, a multiplicity of feedback loops are also traced that quantitatively or qualitatively affect the lives of individuals profiled by the police, while mechanisms such as ghetto classifications and police predictions are fed into governance. This thesis thereby concretely connects the relation between police and the state in the digital era while also accentuating the contradictions of how police power is imagined and enacted. It specifies and details police predictive practices in action, thereby revealing a process that spans the human, non-human and the imaginary
Industrial Excess:Data Storage, Energy and Utility Planning Before, During and After Digital Industrialisation
No full textExcess is usually understood in research as the point at which materiality gets too much. This article shows instead that materiality is always already excessive. The energy utility workers in our study convey that any product making industry also makes excess. In their view, excess as an energetic form is impossible to eliminate from industrial operations. Excess can be reduced, but complete elimination is only possible if industrial operation did not exist. This concretised state of excess becomes apparent when studying the plans facilitating digital industries’ expansion projects. We focus on an implemented utility infrastructure plan for connecting a ‘big tech’ hyperscale datacentre to a public energy system and the classification work it involves. This particular plan leads us to the analytical object of industrial excess. Despite the high impact on public infrastructures and energy consumption, utility plans and these connections with industrialisation projects have been overlooked within scholarship on the digital economy and datacentres. We call this process of connection - making digital industrialisation. Our ethnography with utility workers in Odense, Denmark, shows three analytical entries of boundaries, scales, and admission points into the practices of planning for, with and against excess in connecting expanding industries to publicly owned, non-profitable utility infrastructure. The utility plan shields the energy system against high pollution impacts of digital industrialisation at a municipal scale but exposes its climatic consequences at a transnational scale. The notion of industrial excess devise how forms of industrial product-making and consumptions of industrial products are infrastructurally normalized, and which are not, ultimately giving insight into the radical potential of the non-profitable utility as a figure for ecological transformatio
Addressability: Identification and Communicative Positions in Critical Sociological Perspective
No full textIn this article we develop the concept of addressability to help us unpack processes of identification. We start from a foundational sociological account of addressing as laid out by Simmel, and use Luhmann’s systems theory to identify tensions in the overlap between different systems. The dual character of addressing as reductive (in meaning) and constructive (of communicative positions) helps us understand a mode of knowledge production that generates its own recipients. By concentrating on the moment of addressing in this manner and developing the concept of addressability to explain its complexity, we seek to build an analytic concept that is useful for scholars who are interested in unpacking the construction of communicative positions in identification. We demonstrate the potential of this concept with an analysis of two moments of addressability in action that involve personal identification numbers. We conclude that the intersection and mutual challenge of these two approaches can help us connect different addressing moments while also moving beyond questions of surveillance and entitlement that routinely seek to capture the problem of identification
Ghana's 2024 Elections: Ghanaians Vote for Renewal and Accountability
No full textSince Ghana's return to democratic rule in 1992, the West African country has recurrently been heralded as the model for democracy in Africa. Despite multiple controversies challenging core democratic institutions, Ghana's 2024 elections again represent a strong indicator of the country's democratic resilience. Combining our multi-disciplinary perspectives, we identify the key concerns that preoccupied Ghanaian voters in the lead-up to election day on 7 December 2024. We argue that there is a disconnect between campaign promises, such as the transition into a digital economy, and Ghanaians’ existential worries about the future. Concerns about both environmental and economic liveability equally informed the voter migration behind the 2024 election's unusually large margin of victory. Debates around the alignment of both flagbearers with Ghana's major religious groups, alongside Ghanaians’ rejection of the dismantling of democratic institutions, indicate that Ghana's new government will have to live up to voters’ demands for authenticity and accountability
Modelling Recursion and Probabilistic Choice in Guarded Type Theory
Get PDFConstructive type theory combines logic and programming in one language. This is useful both for reasoning about programs written in type theory, as well as for reasoning about other programming languages inside type theory. It is well-known that it is challenging to extend these applications to languages with recursion and computational effects such as probabilistic choice, because these features are not easily represented in constructive type theory. We show how to define and reason about a programming language with probabilistic choice and recursive types, in guarded type theory. We use higher inductive types to represent finite distributions and guarded recursion to model recursion. We define both operational and denotational semantics, as well as a relation between the two. The relation can be used to prove adequacy, but we also show how to use it to reason about programs up to contextual equivalence
On the Dynamics of Affective States During Play and the Role of Confusion.
Get PDFVideo game designers often view confusion as undesirable, yet it is inevitable, as new players must adapt to new interfaces and mechanics in an increasingly varied and innovative game market, which is more popular than ever. Research suggests that confusion can contribute to a positive experience, potentially motivating players to learn. The state of confusion in video games should be further investigated to gain more insight into the learning experience of play and how it affects the player experience. In this article, we design a study to collect learning-related affects for users playing a game prototype that intentionally confuses the player. We assess the gathered affects against a complex learning model, affirming that, in specific instances, the player experience aligns with the learning experiences. Moreover, we identify correlations between these affects and the Player Experience Inventory constructs, particularly concerning flow experiences
Formant-Based Vowel Categorization for Cross-Lingual Phone Recognition
No full textMultilingual phone recognition models can learn language-independent pronunciation patterns from large volumes of spoken data and recognize them across languages. This potential can be harnessed to improve speech technologies for underresourced languages. However, these models are typically trained on phonological representations of speech sounds, which do not necessarily reflect the phonetic realization of speech. A mismatch between a phonological symbol and its phonetic realizations can lead to phone confusions and reduce performance. This work introduces formant-based vowel categorization aimed at improving cross-lingual vowel recognition by uncovering a vowel's phonetic quality from its formant frequencies, and reorganizing the vowel categories in a multilingual speech corpus to increase their consistency across languages. The work investigates vowel categories obtained from a trilingual multi-dialect speech corpus of Danish, Norwegian, and Swedish using three categorization techniques. Cross-lingual phone recognition experiments reveal that uniting vowel categories of different languages into a set of shared formant-based categories improves cross-lingual recognition of the shared vowels, but also interferes with recognition of vowels not present in one or more training languages. Cross-lingual evaluation on regional dialects provides inconclusive results. Nevertheless, improved recognition of individual vowels can translate to improvements in overall phone recognition on languages unseen during training
Synthetic emotions and Gamification: Exploring Engagement Strategies for Small Social Robots in different age-groups
No full textThis paper investigates engagement strategies for pocket-sized social robots designed to aid children in the longer term with anxiety disorders. We explore engagement strategies for a pocket-sized tactile robot designed to support children with anxiety through daily interaction. The robot delivers either synthetic emotional feedback or gamified point rewards to encourage user participation. We evaluated these strategies through two studies conducted in contrasting age groups and settings: a school-based preference study with young children, and a full-day behavioral study with university students in naturalistic environments. The study with school children (aged 6-8, n=16) indicated a preference for emotional engagement over points-based approaches. The follow up study with university students (n=14) across a full day of interactions revealed contrasting results: points-based systems produced significantly higher task accuracy (p < 0.05) and sustained performance over time. Points-based participants also rated their robot as more human-like and likable, challenging assumptions about emotional expressiveness being necessary for positive robot perception. These findings across different age groups suggest that stated preferences may not predict behavioral effectiveness, with structured reward mechanisms demonstrating superior performance outcomes despite lower user preference ratings in younger populations. This work contributes insights into age-related differences in engagement strategy effectiveness in human-robot interaction design, highlighting the importance of behavioral validation alongside preference assessment in therapeutic robotics