The processing of textual data to obtain psychological measures is becoming prevalent. However, the problem is that in most cases, tagged corpora are unavailable and therefore conventional classification methods cannot be easily applied. We present a novel solution for measuring psychological dimensions in textual data. The solution is based on (i) the construction of patterns from validated psychological questionnaires measuring the target dimension and (ii) similarity measurements of some of the text's sentences to the pre-defined patterns. Our proposed methodology was tested on (i) the identification of the five personality dimensions in a gold standard corpus, (ii) the measurement of depression and (iii) the screening of civilian mass murderers, gaining preliminary support for its validity.

Full natural language understanding requires identifying and analyzing the meanings of metaphors, which are ubiquitous in both text and speech. Over the last thirty years, linguistic metaphors have been shown to be based on more general conceptual metaphors, partial semantic mappings between disparate conceptual domains. Though some achievements have been made in identifying linguistic metaphors over the last decade or so, little work has been done to date on automatically identifying conceptual metaphors. This paper describes research on identifying conceptual metaphors based on corpus data. Our method uses as little background knowledge as possible, to ease transfer to new languages and to mini- mize any bias introduced by the knowledge base construction process. The method relies on general heuristics for identifying linguistic metaphors and statistical clustering (guided by Wordnet) to form conceptual metaphor candidates. Human experiments show the system effectively finds meaningful conceptual metaphors.

In his seminal and pioneering work on the social context of science, Ludwick Fleck [2] argues that ideas acquire “magical power” simply by being used. “Information” is a concept that has acquired magical power despite the fact that its specific context of use—Shannon's theory of communication—was far removed from various fields in which it has been applied, such as biology. In retrospect, one cannot separate the magical power information has acquired from the revolutionary success of information technology and its widespread use. Nevertheless, the meaning of information and its relevance as an organizing concept for other fields of inquiry, such as biology, is far from clear. In this context, examining the meaning and role of information in living systems is a welcomed venture.The current book has 13 chapters organized around (1) the definition of life, (2) information and biological organization, and (3) information and the biology of cognition, value, and language. As it is impossible to review all the chapters, I have focused on three representative examples that reflect my personal interest.The opening chapter by Ruiz-Mirazo and Moreno deals with the definition of life. The authors criticize the reductionist approach, arguing that it overshadowed the “idea that there is something else relevant for biology to do, namely to explain the complex dynamic integration of those components into functional units.” This awareness is clearly in line with the non-reductionist and systemic approach to biology (e.g., [1, 5, 7]). In this context, what should a definition of life look like? That is a tricky issue, as I have recently argued in response to a novel definition [6]. The authors are well aware of the difficulty in dealing with a definition of life. Interestingly, they suggest that in order to define life, what matters “is determining the set of enabling constraints on the physical and chemical laws.” This proposal clearly echoes Michael Polanyi's seminal article and avoids the essentialist fallacy they criticize. Here we get into the authors' tentative definition of life, which involves two interacting and complementary levels of analysis: the individual autonomous agent and the meta-network of these agents, which generate information that is fed back and forth between the levels. This idea echoes Gregory Bateson's idea of a recursive hierarchy [3] and his idea of information as a relational construct. The challenge facing the authors in defining life along these lines is to identify the precise nature of the constraints operating on living systems and to discuss them in terms of information in order to better describe their role in constituting the organization of living systems and their landscape of potential complexity.In the second part of the book, Queiroz, Emmeche, Kull, and El-Hani present the biosemiotic approach to biology. The appearance of a chapter dealing with biosemiotics in a book dedicated to information is far from trivial. From the perspective of information theory, living systems are just one concrete realization of a Turing machine that processes strings of letters. For the semiotician, in contrast, the basic unit of analysis is the sign that has a totally different sense from a bit. From the biosemiotic perspective (e.g., [4]), living systems process signs rather than information. The authors draw on a specific semiotic perspective, the one developed by Charles Sanders Peirce, to describe the genetic system and signal transduction in the immune system. They present the argument that the probabilistic measure of information is indifferent to meaning, whereas the way living systems make sense of signals is crucial for understanding their behavior. In contrast to information theory, the idea of meaning is built into Peirce's triadic definition of the sign. According to this thesis, a sign, whether a biological sign or a linguistic sign, is a non-reducible triad composed of the sign, the signified, and the interpretant—the effect the sign produces on the interpreter. Applying this perspective to the genetic system, we learn that the process of sign use—semiosis—is actually information and that in the DNA we can find information only “in potency.” Genes are signs, but without their transcription and their translation into proteins, they have no meaning and hence no information value. This biosemiotic perspective has an appealing descriptive value. However, in order to present a serious alternative to the information theory perspective, biosemiotics, like bioinformatics, should provide rigorous evidence supporting its theoretical and explanatory value. Bioinformatics has already passed this test, while biosemiotics, as a young and radical venture, is still striving to establish itself as an appealing alternative for the working scientist.The idea of constraints that we have already encountered in the first chapter returns in the third part of the book, in a chapter about information theory and perception written by Baddeley, Vincent, and Attewell. The authors suggest that early vision is amenable to study using information theory for several reasons, which include the amount of data gathered by researchers with regard to the process of early vision. The transmission of information from the eyes to the cortex involves several bottlenecks. Here the importance of constraints enters the picture, as “A system that is under severe constraints about how much information can be transmitted is more likely to be, in some sense, efficient” (p. 290). For instance, the image formed on the retina cannot be transmitted to the brain as it is. Therefore, the images transmitted to the brain are compressed, and this process of compression can be studied through information theory. Another constraint is the energy consumption of neurons in the brain. Therefore, the authors present a model that maximizes information transmission while minimizing energy consumption. The model they present and its empirical support are described as a “major breakthrough” in the study of vision. Indeed, this model is an excellent illustration of the way information theory may be used to model the behavior of living systems by taking into account real-world constraints.The three chapters here presented as illustrative examples represent the way information is a relevant theoretical construct for the study of living systems, a construct that may be used to develop new conceptualization about big questions facing biology (e.g., the definition of life), that may be formulated in novel and radical ways (e.g., information as semiosis), and that may be used to model real-world constraints in order to better understand the behavior of living systems, as in the case of early vision.In sum, this edited book introduces a polyphonic view of information and its relevance for understanding living systems, and therefore it is a promising venture.

Representing meaning is a major challenge facing Web 3.0. However, it is extremely difficult to excavate the meaning of a target concept from textual data as there is no one-to-one correspondence between the textual unit in which the target concept is embedded and the conceptual content that we would like to excavate. In this paper, we propose one possible approach for addressing this challenge, by harvesting the Web for metaphorical relations in which the target concept is an argument. In this paper, we present several preliminary results of "Pedesis" - a novel automated system for excavating the meaning of target concepts - and plan to present at the oral presentation an application of this methodology for identifying depression in free text.