In this paper, a new approach to weakly supervised semantic segmentation is proposed. The main goal in semantic segmentation is to assign a semantic label to each pixel. In weakly supervised setting, each training image is only labeled by the classes they contain, not by their locations. The main contribution of this paper is to simultaneously incorporate the Object level and context level information in assigning class label to each pixel of the image. To do this, regions in each image are grouped such that groups of regions in images with the same semantic label have the same appearance and context. To do this, an iterative move-making algorithm is proposed. At first, each pixel of the image is initially labeled and then model of appearance and context for each class label is learned. Then, semantic label of each pixel is updated such that the regions with the same sematic label have the same appearance and context in the set of images. In the next step, appearance and context models for each semantic class are updated. It is repeated until in the two consecutive epochs, labels of the pixels are not changed. To evaluate our proposed approach, it is applied on the MSRC dataset. The obtained results show that our approach outperforms comparable state-of-the-art approaches.

Purpose: This paper presents the results of research on the information science evolution, its concepts and problems, highlighting how the recent post-truth phenomenon raises new questions to the area. The Objectives are related to the identification and characterization of the post-truth phenomenon to demonstrate that it represents a new informational reality challenging the categories hitherto existing. The discussion on information science starts from the mapping of three major study models in the area: the physical, the cognitive, and the social. The first, which arose at the beginning of information science, focuses on the technical aspects of the transmission of records and on issues of information retrieval. The second focuses on the design of information systems based on the users' cognitive needs and search processes. The third focuses on the social and cultural dimensions of information, its relationship with the action of the subjects and the contexts in which these actions occur. The discussion about the post-truth is presented in terms of its causes, characteristics and consequences. Its causes include scientific denialism, the human cognitive bias of confirmation, the disintermediation of information, the formation of the bubble effect as a result of the logic of the algorithms of social networks and search engines, and the hijacking of the idea of relativizing the truth produced by the philosophical movement of postmodernity. Regarding the characteristics of the phenomenon, the intentional and spontaneous forms of disinformation production stand out, through different platforms, mainly the instant messaging applications that work underground. Finally, in relation to the consequences, the strengthening of authoritarian regimes, the weakening of institutions and a greater submission of subjects are listed. Method: The methodology used in this research is literature review and epistemological discussion. For information science, books, manuals and articles produced by authors from different countries were considered. Their Objective was to present or map the field from its theoretical currents and paradigms. Thus, for the analysis, works that only made a history of information science or those presented it only from a single point of view or theoretical perspective were not considered. For the question of post-truth, mainly recent books were considered for the depth of discussions and general analyses of the phenomenon and exploring its causes, characteristics and consequences. Analyses that correlated post-truth with other contemporary phenomena were also considered. The works selected to compose the theoretical framework were analyzed from an epistemological viewpoint in order to compose an interpretive framework to meet the proposed Objectives. Findings: The result of the confrontation between the discussion of the conceptual framework of information science and the questions related to post-truth showed the need for information science to give centrality to the attribute of "truth" of information, in addition to those already studied, such as relevance, retrieval, relationship with knowledge. The phenomenon of post-truth has technical, human, and social dimensions. information science has the tools and concepts to explore these dimensions. Conclusion: Post-truth phenomenon presents itself as the major contemporary challenge for information science, requiring further investigation for understanding it, and developing strategies to deal with it.

In this paper, a new mechanism is proposed to transform the structural modeling elements of the UML class diagram and Object-Z specifications into each other. A set of bidirectional rules is defined to transform the mentioned elements into each other. Bidirectional transformation of the UML class diagram, as one of the most useful diagrams of UML, and Object-Z specifications into each other prepares the ground for the use of the unique advantages of both formal and visual modeling methods. The feasibility of the proposed approach is evaluated using the multi-lift case study. The results of conducting the multi-lift case study show that the proposed mechanism is feasible.

Languages employ two language tools to mark the Objects which are case marking and agreement. The present study is intended to investigate agreement as an important aspect of DOM in Kalhori Kurdish. To this end, information structure as a functional approach was adopted as the theoretical basis of the study (Lambrecht, 1994, 2000; Dalrymple & Nikoleava, 2010, Nikoleava, 2001). This study is a descriptive-analytical one in which the data were collected through interview, observation, and consultation with authentic library sources. The results indicated that in Kalhori Kurdish, Object agreement is performed as a representation DOM with Objects that have verbal prominence and their instance has already been mentioned in the discourse. This agreement is in terms of person and number which is done by a group of Objective enclitics appeared on the verb. Another finding was that in Kalhori Kurdish, subject agreement is not sensitive to information structure. Furthermore, it was found that indirect Objects can not trigger Object agreement, even if they are topical. Finally, it is possible to delete topical Objects and subjects or to postpose them to post-verbal position, provided that they are topical.

Direct Objects are encoded via specific grammatical tools in Persian. Using random data mostly from colloquial Persian (and written Persian in case of need), the author first intend to introduce these tools. Employing the Layered Structure of the Clause and Focus Structure as defined in Role and Reference Grammar, the author proceeds to formally represent the tools via syntactic templates that contain information-structural properties of the direct Object. We learn that the hypothesis regarding the direct Object being a secondary topic marker in the spirit of ra-marking (Dabir-Moghaddam, 2006) needs to be discarded. information-structurally, not only is the direct Object qualified to be part of the pragmatic presupposition, hence serving as the topic of the sentence, it is also licensed to constitute a narrow focus by itself or to participate in predicate and sentence focus structures. This originates from the fact that-ra in Persian signals the identifiability of the Object’ s referent in the interlocutors’ discourse universe. It is also proposed that the bi-fold classification of a narrow focus into in-situ information focus and ex-situ contrastive focus does not only apply to the narrow focus in the spirit of Persian data. A predicate focus can be contrastive when a ra-marked Object residing at the Pre-Core Slot is disjoined from the transitive predicate by an intervening subject. In actuality, this discontinuous predicate focus structure is ‘ construed’ as a contrastive narrow focus.

SLAM Loop Closure Detection information Theory Kolmogorov Complexity In this paper the problem of 3D scene and Object classification from depth data is addressed. In contrast to high-dimensional feature-based representation, the depth data is described in a low dimensional space. In order to remedy the curse of dimensionality problem, the depth data is described by a sparse model over a learned dictionary. Exploiting the algorithmic information theory, a new definition for the Kolmogorov complexity is presented based on the Earth Mover’s Distance (EMD). Finally the classification of 3D scenes and Objects is accomplished by means of a normalized complexity distance, where its applicability in practice is proved by some experiments on publicly available datasets. Also, the experimental results are compared to some state-of-the-art 3D Object classification methods. Furthermore, it has been shown that the proposed method outperforms FAB-Map 2.0 in detecting loop closures, in the sense of the precision and recall.

Human visual system operates superior than best machine vision systems in Object recognition. So, researchers in machine vision and neuroscience try to model human visual system in order to employ it in machines. HMAX is one of the best operating models in this area. It is based on the function of brain cells in the ventral stream of visual cortex and contains four computational layers. In the learning stage, many image partitions called image patches are extracted randomly with different sizes from training images. This random selection of image patches is one of the drawbacks of HMAX which decreases the performance and increases the computational complexity of the algorithm. In this paper, a novel patch selection from the set of random patches is proposed. In this method, using a recursive approach, optimal patches are selected from optimal features of training images by mutual information maximization feature selection. The performance of proposed algorithm in binary classification (existence or non-existence of Objects in the images) is compared with HMAX and the superiority is proved.