Search

The document clustering technique is widely recognized as a useful tool for information retrieval, organizing web documents, text mining and so on. The purpose of this paper is to review various document clustering techniques, and to discuss research issues for enhancing effectiveness or efficiency of the clustering methods. We explore extensive literature on non-hierarchical methods (single-pass methods), hierarchical methods (single-link, complete-link, etc.), dimensional reduction methods (LSI, principal component analysis, etc.), probabilistic methods, data mining techniques, and so on. In particular, this paper focuses on typical techniques, such as the k-means algorithm, the leader-follower algorithm, self-organizing map (SOM). single- or complete-link methods, bisecting k-means methods, latent semantic indexing (LSI), Gaussian-Mixture model and so on. After reviewing the techniques and algorithms, we discuss research issues on document clustering: computational complexity, feature extraction (selection of words), methods for defining term weights and similarity, and evaluation of results. [ABSTRACT FROM AUTHOR]

That artificial intelligence (AI), algorithms, and related technologies could use a few good booster shots of “humanism” is widely apparent. In both program code and implementation, AI and algorithms have been accused of harbouring deep-seated flaws that conflict with human values. They are prime examples of the skew towards white, Western, men and demonstrate the bankruptcy in the face of neoliberalist, profit- and market-oriented social paradigms that this special issue seeks to address. Currently, computer scientists and AI researchers who are looking to remedy these problems are often in favour of more data, more powerful machines, more complex algorithms—in short, that we should fix problems with AI by building better AI. In this view human beings and the world can be modelled in code—our lives, interactions, society, and our very selves can be broken down into data points which can be assessed by highly advanced technologies. When these scientists and researchers seek to broaden their approach they often look to philosophy. However, the philosophy they look to is overwhelmingly Anglo-analytic, which views the world in extremely similar ways. Both AI and Anglo-analytic philosophy argue for solutions to humanistic problems which are essentially mathematical. They share in seeing important concepts, such as persons, emotions, agency, and ethics, as mechanistic, atomistic, and calculable. In this paper I will argue that Classical Chinese philosophy offer insightful resources for addressing the humanist problems in AI. Rather than arguing for mathematical solutions, or envisioning persons, emotions, agency, and ethics, as other rigid, atomistic, and mechanistic approaches, Chinese philosophy emphasizes transformation, interrelatedness, and correlative developments. Accordingly, it offers tools for appreciating the world, society, and ourselves as spontaneous, complex, and full of tension. AI can be programmed and used in ways that do not reduce the complexity and conflict in the world, but provide us instead with tools to make sense of it—tools that are humanistic in nature. To this end, Chinese philosophy can be a helpful collaborative partner.