Browsing by Author "Brand, Dirk"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Algorithmic decision-making and the law
    (Department for E-Governance and Administration, Danube University Krems, 2020) Brand, Dirk
    The Fourth Industrial Revolution is reshaping the world we know dramatically and is characterised by a close interaction between the biological, digital and physical spheres. Digital technologies are impacting all facets of our lives and create a series of new opportunities but also various challenges. The Fourth Industrial Revolution does not follow a linear development trajectory, but due to the diverse nature and rapid pace of technological developments, could rather be compared to a series of networks with multiple connecting points. This has caused the development of the law which deals with these concerns to generally be slow and unable to match the pace and scope of technological developments. In the context of public law there are many questions and challenges relating to individual rights, for example the right to privacy, and the role and responsibilities of government relating to policy development and regulation dealing with the Fourth Industrial Revolution. The concept of a Rechtsstaat could arguably provide an appropriate legal framework for shaping the ethical framework, normative standards and a value-based governance model for the Fourth Industrial Revolution, including for algorithmic decision-making. The public law concept of accountability should be contextualised in order to apply it to algorithmic decision-making. In the data-driven economy of the 21st century the pace and scope of technological developments that impact humanity requires the development of appropriate legal frameworks to reflect and accommodate the needs of society, in particular relating to the recognition of fundamental human rights. It is concluded that a broad set of ethical and legal principles, which can guide the development of international and national legal frameworks to regulate algorithmic decision-making, is needed.
  • Thumbnail Image
    Item
    Comment classification for an online news domain
    (2014-12) Brand, Dirk; Van der Merwe, Brink
    ENGLISH ABSTRACT: In online discussion forums, comment moderation systems are often faced with the problem of establishing the value of an unseen online comment. By knowing the value of comments, the system is empowered to establish rank and to enhance the user experience. It is also useful for identifying malicious users that consistently show behaviour that is detrimental to the community. In this paper, we investigate and evaluate various machine learning techniques for automatic comment scoring. We derive a set of features that aim to capture various comment quality metrics (like relevance, informativeness and spelling) and compare it to content-based features. We investigate the correlation of these features against the community popularity of the comments. Through investigation of supervised learning techniques, we show that content-based features better serves as a predictor of popularity, while quality-based features are better suited for predicting user engagement. We also evaluate how well our classifier based rankings correlate to community preference.
  • Thumbnail Image
    Item
    Local government finance : a comparative study
    (SUN MeDIA Stellenbosch, 2016) Brand, Dirk
    ENGLISH SUMMARY : Introduction: In a multi-sphere system of government, such as the constitutional system in South Africa which is characterised by cooperative government, effective financial intergovernmental relations are essential for the effective functioning of the whole constitutional system. Financial constitutional law inter alia includes the actual allocation of functions and financial resources to the three spheres of government, equalisation measures, policy issues relating to financial intergovernmental relations, as well as governance issues. After two decades of constitutional democracy in South Africa, various aspects of the financial intergovernmental relations system are in the spotlight. Annual reports from the Auditor-General, regular newspaper reports and political debates in different legislatures confirm the bad state of affairs in many municipalities, various provinces and national government departments. It is in local government in particular that the consequences of a range of problems relating to the finances of municipalities are often quite visible since it translates into bad or no service delivery. This publication is thus focused on the local government finance model, but reference will also be made to the rest of the system of financial intergovernmental relations where relevant.
  • Thumbnail Image
    Item
    N-gram representations for comment filtering
    (ACM, Inc., 2015-09) Brand, Dirk; Kroon, Steve; Van der Merwe, Brink; Cleophas, Loek
    Accurate classifiers for short texts are valuable assets in many applications. Especially in online communities, where users contribute to content in the form of posts and comments, an effective way of automatically categorising posts proves highly valuable. This paper investigates the use of N- grams as features for short text classification, and compares it to manual feature design techniques that have been popu- lar in this domain. We find that the N-gram representations greatly outperform manual feature extraction techniques.
  • Thumbnail Image
    Item
    Sample evaluation for action selection in Monte Carlo Tree Search
    (2014) Brand, Dirk; Kroon, Steve
    ENGLISH ABSTRACT; Building sophisticated computer players for games has been of interest since the advent of artificial intelligence research. Monte Carlo tree search (MCTS) techniques have led to recent advances in the performance of computer players in a variety of games. Without any refinements, the commonly used upper confidence bounds applied to trees (UCT) selection policy for MCTS performs poorly on games with high branching factors, because an inordinate amount of time is spent performing simulations from each sibling of a node before that node can be further investigated. Move-ordering heuristics are usually proposed to address this issue, but when the branching factor is large, it can be costly to order candidate actions. We propose a technique combining sampling from the action space with a naive evaluation function for identifying nodes to add to the tree when using MCTS in cases where the branching factor is large. The approach is evaluated on a restricted version of the board game Risk with promising results.