Search

In this paper, we propose to use a convertible deep neural network (DNN) model with a transfer adaptation mechanism to deal with varying input and output numbers of neurons. The flexible DNN model serves as a multi-label classifier for the recommender system as part of the retrieval systems' push mechanism, which learns the combination of tabular features and proposes the number of discrete offers (targets). Our retrieval system uses the transfer adaptation, mechanism, when the number of features changes, it replaces the input layer of the neural network then freezes all gradients on the following layers, trains only replaced layer, and unfreezes the entire model. The experiments show that using the transfer adaptation technique impacts stable loss decreasing and learning speed during the training process. Furthermore, our proposed model demonstrates notable advantages in production scenarios. Specifically, it exhibits enhanced efficiency, manifesting in accelerated processing times and improved resource utilization, thereby contributing to a more sustainable and cost-effective training of machine learning solutions in real-world applications. [ABSTRACT FROM AUTHOR]

Robot swarms have the potential to be used as an out-of-the-box solution for storage and retrieval that is low cost, scalable to the needs of the task, and would require minimal set up and training for the users. Swarms are adaptable, robust and scalable with a relatively low computational cost which makes them appropriate for this purpose. This project simulated a robot swarm with simple sensors and stochastic movement, collecting boxes from storage to deliver them to the user. We show in simulation that stochastic strategies based on random walk and probabilistic sampling of local boxes could give rise to competitive solutions to retrieve boxes and deliver them unordered, or following a predetermined order, within a storage scenario. The performance of the task is drastically improved using an additional simple bias rule which uses compass measurements and does not reduce the minimalism of the control. It is shown that swarm technology could provide an out-of-the-box system for storage and retrieval using only information local to each robot and with distributed control. [ABSTRACT FROM AUTHOR]

We discuss the problem of people retrieval by means of composite pictures and methods of its practical realization. Earlier on, the problem was posed in the previous paper by the authors, and this paper deals with its further development. The starting premise here is that for the successful search of people by their sketches it is necessary to transform these sketches into sets of populations of sketches imitating evidence of «group of witnesses» and evidence with incomplete information in verbal portraits. Variants of structures for benchmark «photo-sketch» databases are presented, intended for modeling and practical realization of original photos retrieval by sketches, which new component is a population of sketches. Problems of preprocessing for initial sketches and original photos and its influence on the result of their comparison are discussed. Simple sketch recognition systems (Simple FaRetSys) and a problem of original photos retrieval by the sketches are considered. Shortcomings of such systems are shown and new decisions on extending and development of simple systems (Extended FaRetSys) are presented. Experiments on searching of original photos by sketches in the CUFS database of sketches and similar experiments on widely known FERET and CUFSF facial databases are presented. Three frameworks are offered for retrieval performance improvement. In the first one, original sketches are transformed into populations, and then in these populations the sketch similar to the given sketch (Forensic Sketch) is already defined. The class of the sketch found in a population «by definition» unambiguously corresponds to a class of the original photo. In the second framework, the Forensic Sketch is transformed to a population of sketches, and all original sketches in a benchmarking database are compared to sketches from populations of the Forensic Sketch. The class of matches is determined in the same manner as in the first framework. The third framework includes generation of a population of sketches, both from all original sketches, and from all Forensic Sketches. The further line of research is obvious: retrieval by matching between sketches of these two populations.

OBJECTIVE To investigate the awareness and use of online information retrieval systems by health professionals in clinical practice. METHOD A cross-sectional study was performed with a sample of 439 physicians and nurses from primary and secondary health care, using a questionnaire measuring retrieval behavior, utilization of online evidence, and reasons for and barriers to information retrieval. Data collection took place, simultaneously, online and by printed distribution of the instrument. RESULTS The mean age of the respondents was 38.9 years and the mean duration of their work was 12.6 years. Health professionals with experience in using computers and a high level of education appeared to have a greater likelihood of using bibliographic databases (p<0.027). Participants with postgraduate studies more often reported finding the information they were looking for in electronic databases (p<0.001). The main obstacles to using bibliographic databases were lack of access and the culture of using other information sources. CONCLUSIONS More technical, educational and organizational support in the use of electronic evidence databases is needed for physicians and nurses, to promote elicitation of knowledge and information with a view to improving the patient care that they provide. [ABSTRACT FROM AUTHOR]

Displays of art in public or private spaces have long been of interest to curators, gallerists, artists and art historians. The emergence of gallery paintings at the beginning of the seventeenth century and the photographic documentation of (modern) exhibitions testify to that. Taken as factual documents, these images are not only representative of social status, wealth or the museum’s thematic focus, but also contain information about artistic relations and exhibition practices. Digitization efforts of previous years have made these documents, including photographs, catalogs or press releases, available to public audiences and scholars. While a manual analysis has proved to be insufficient, because of the sheer number of available data, computational approaches and tools allowed for a greater access. The following article describes how digital images of exhibitions, as released by the New York Museum of Modern Art in the fall of 2016, are studied with a retrieval system to analyze in which artistic contexts selected artworks were presented in exhibits.

In the paper we study how the choice of color space affects the user-friendliness of the graphical user interface of CBIR. The pilot study was carried out on 46 students of the Department of Library and Information Science and Book Studies at the Faculty of Arts at the University of Ljubljana. The study showed that user interfaces based on HSV color space are simple to use even for users who do not possess the theoretical knowledge about colors and do not understand HSV color space, while even reasonably good understanding of RGB color space does not guarantee frustration-free user experience when using RGB user interface. User interfaces offering color choice just by clicking were better received than user interfaces that make use of color sliders. This article provides guidance for designing an interface for selecting color in CBIR systems. [ABSTRACT FROM AUTHOR]

The content-based retrieval systems for 3D models on the Web become necessary since digital databases of 3D objects are growing. In this paper, we propose a new method to describe 3D models. This method is based on 3D discrete cosine transform which is applied for the voxelized 3D model. The discrete cosine transform is widely used for 2D image compress ion and it shows its performance for the JPEG compression algorithm. The proposed descriptor is implemented in our 3D search engine, tested using the princeton shape benchmark database, tested for noise and decimation; and compared to other 3D descriptors known in the literature. [ABSTRACT FROM AUTHOR]

Three-dimensional model databases used around the Web become larger; therefore, the development of retrieval systems for such databases is required. In this paper, we propose a new 3D shape descriptor and a web based search engine for 3D models given by polygonal meshes. The main idea is to reconstruct a 3D closed curve that represents the 3D object and to extract feature vectors from it. We combine two new descriptors the area and the dot product descriptors that describe the reconstructed 3D closed curve in order to define the 3D curve analysis descriptor. The proposed method is robust for noise, compared to other descriptors known in the literature and tested using the Princeton shape benchmark database via the search engine. [ABSTRACT FROM AUTHOR]

Information systems (IS) project failure is a costly and common problem. This is despite advances in development tools and technologies. In this article, we argue that one reason for this is the failure of project post-mortems to generate constructive “lessons learned” from previous projects. Over time, the ineffective practices would persist in the organization, rendering it resistant to change. The attribution theory literature serves as one of the few promising theoretical bases to explain why project post-mortems fail. A case study of a project post-mortem undertaken for an abandoned electronic procurement system project is discussed and analyzed. We identify five antecedent conditions of attribution error: the presence of self-appointed mindsets, the general persistence of negative beliefs, memory decay, selective recall of project events and the influence of power dynamics within the organization. We discuss the research and practical implications of these findings and suggest how the problem of attribution error may be minimized in project post-mortems. [ABSTRACT FROM AUTHOR]

With development of technology, large image databases have become reality. Pictures in these databases were mainly described with keywords, but due to insufficiency of such description, the need for different approach to image indexing and retrieval has arisen. Content-based image retrieval is an alternative to textual based systems. Images are indexed on the basis of their actual content, features like color, texture and shape are presented numerically. In retrieval process, features for input image are extracted with the same algorithm as used for pictures already stored in the database, and values are compared. Such systems are known as CBIR (content-based image retrieval) systems. Basic characteristics of CBIR systems, which are freely available on the Internet, are presented. The systems are compared with respect to the environment in which they were built, applied features for content description and ability of user interface. [ABSTRACT FROM AUTHOR]