We consider the problem of modeling annotated data---data with multiple types where the instance of one type (such as a caption) serves as a description of the other type (such as an image). We describe three hierarchical probabilistic mixture models which aim to describe such data, culminating in correspondence latent Dirichlet allocation, a latent variable model that is effective at modeling the joint distribution of both types and the conditional distribution of the annotation given the primary type. We conduct experiments on the Corel database of images and captions, assessing performance in terms of held-out likelihood, automatic annotation, and text-based image retrieval.

/pdf/modeling-annotated-data-3cdvhscvlq.pdf

Modeling annotated data

13 EADERS of Mr. Bertrand Russell's philosophical £v works know that one of his pupils before the outbreak of the war, an Austrian, Mr. Ludwig Wittgenstein, caused him to change his views in some important particulars. Curiosity can now be satisfied. The “Tractatus Logico-Philosophicus “which Mr. Ogden has included in his new library of philosophy is a remarkable and strikingly original work. It is published in German and English in parallel pages. It is difficult to appreciate the reason for this, seeing that the author is evidently familiar with our language and has himself carefully revised the proofs of the translation. Also we should have liked to have the Tractatus without Mr. Russell's Introduction, not, we hasten to add, on account of any fault or shortcoming in that introduction, which is highly appreciative and in part a defence of himself, in part explanatory of the author, but for the reason that good wine needs no bush and that Mr. Russell's bush has the unfortunate effect of dulling the palate instead of whetting the appetite. In his penultimate sentence Mr. Russell says; “To have constructed a theory of logic which is not at any point obviously wrong is to have achieved a work of extraordinary difficulty and importance.” We agree, but how uninspiring when compared with Mr. Wittgenstein's own statement of aim: “What can be said at all can be said clearly, and whereof one cannot speak, thereof one must be silent.” Tractatus Logico-Philosophicus. By Ludwig Wittgenstein. (International Library of Psychology, Philosophy and Scientific Method.) Pp. 189. (London: Kegan Paul and Co., Ltd.; New York: Harcourt, Brace and Co., Inc., 1922.) 10s. 6d. net.

Tractatus Logico-Philosophicus

In the era of the Internet of Things (IoT), an enormous amount of sensing devices collect and/or generate various sensory data over time for a wide range of fields and applications. Based on the nature of the application, these devices will result in big or fast/real-time data streams. Applying analytics over such data streams to discover new information, predict future insights, and make control decisions is a crucial process that makes IoT a worthy paradigm for businesses and a quality-of-life improving technology. In this paper, we provide a thorough overview on using a class of advanced machine learning techniques, namely deep learning (DL), to facilitate the analytics and learning in the IoT domain. We start by articulating IoT data characteristics and identifying two major treatments for IoT data from a machine learning perspective, namely IoT big data analytics and IoT streaming data analytics. We also discuss why DL is a promising approach to achieve the desired analytics in these types of data and applications. The potential of using emerging DL techniques for IoT data analytics are then discussed, and its promises and challenges are introduced. We present a comprehensive background on different DL architectures and algorithms. We also analyze and summarize major reported research attempts that leveraged DL in the IoT domain. The smart IoT devices that have incorporated DL in their intelligence background are also discussed. DL implementation approaches on the fog and cloud centers in support of IoT applications are also surveyed. Finally, we shed light on some challenges and potential directions for future research. At the end of each section, we highlight the lessons learned based on our experiments and review of the recent literature.

Deep Learning for IoT Big Data and Streaming Analytics: A Survey

The notion of fuzziness as defined in this paper relates to situations in which the source of imprecision is not a random variable or a stochastic process, but rather a class or classes which do not possess sharply defined boundaries, e.g., the “class of bald men,” or the “class of numbers which are much greater than 10,” or the “class of adaptive systems,” etc. A basic concept which makes it possible to treat fuzziness in a quantitative manner is that of a fuzzy set, that is, a class in which there may be grades of membership intermediate between full membership and non-membership. Thus, a fuzzy set is characterized by a membership function which assigns to each object its grade of membership (a number lying between 0 and 1) in the fuzzy set. After a review of some of the relevant properties of fuzzy sets, the notions of a fuzzy system and a fuzzy class of systems are introduced and briefly analyzed. The paper closes with a section dealing with optimization under fuzzy constraints in which an approach to...

Fuzzy sets and systems

Purpose – The purpose of this paper is to identify and describe the most prominent research areas connected with “Big Data” and propose a thorough definition of the term. Design/methodology/approach – The authors have analysed a conspicuous corpus of industry and academia articles linked with Big Data to find commonalities among the topics they treated. The authors have also compiled a survey of existing definitions with a view of generating a more solid one that encompasses most of the work happening in the field. Findings – The main themes of Big Data are: information, technology, methods and impact. The authors propose a new definition for the term that reads as follows: “Big Data is the Information asset characterized by such a High Volume, Velocity and Variety to require specific Technology and Analytical Methods for its transformation into Value.” Practical implications – The formal definition that is proposed can enable a more coherent development of the concept of Big Data, as it solely relies on ...

A formal definition of Big Data based on its essential features

We propose an effective CNN architecture for visual parking occupancy detectionThe CNN architecture is small enough to run on smart camerasThe proposed solution performs and generalizes better than other SotA approachesWe provide a new training/validation dataset for parking occupancy detection A smart camera is a vision system capable of extracting application-specific information from the captured images The paper proposes a decentralized and efficient solution for visual parking lot occupancy detection based on a deep Convolutional Neural Network (CNN) specifically designed for smart cameras This solution is compared with state-of-the-art approaches using two visual datasets: PKLot, already existing in literature, and CNRPark-EXT The former is an existing dataset, that allowed us to exhaustively compare with previous works The latter dataset has been created in the context of this research, accumulating data across various seasons of the year, to test our approach in particularly challenging situations, exhibiting occlusions, and diverse and difficult viewpoints This dataset is public available to the scientific community and is another contribution of our research Our experiments show that our solution outperforms and generalizes the best performing approaches on both datasets The performance of our proposed CNN architecture on the parking lot occupancy detection task, is comparable to the well-known AlexNet, which is three orders of magnitude larger

Deep learning for decentralized parking lot occupancy detection

The Digital Library universe is a complex framework. The growth and evolution of this framework in terms of approaches, solutions and systems has led to the need for common foundations capable of setting the basis for better understanding, communicating and stimulating further evolution in this area. The DELOS Digital Library Reference Model aims at contributing to the creation of such foundations. It exploits the collective understanding on Digital Libraries that has been acquired by European research groups active in the Digital Library field for many years, both within the DELOS Network of Excellence and beyond, as well as by other groups around the world. It identifies the set of concepts and relationships that characterise the essence of the Digital Library universe. This model should be considered as a roadmap allowing the various players involved in the Digital Library domain to follow the same route and share a common understanding in dealing with the entities of such a universe. This volume presents the DELOS Reference Model by introducing the principles governing it as well as the set of concepts and relationships that collectively capture the intrinsic nature of the various entities of the Digital Library universe. Because of the broad coverage of the Digital Library universe, its evolving nature, and the lack of any previous agreement on its foundations, the Reference Model is by necessity a dynamic framework, as is also this document. Continuous evolutions of the document are envisaged in order to obtain a number of well-formed and consolidated definitions, shared by the Digital Library community.

The Delos digital library reference model : foundations for digital libraries

According to the logical model of Information Retrieval (IR), the task of IR can be described as the extraction, from a given document base, of those documents d that, given a query q, make the formula d ! q valid, where d and q are formulae of the chosen logic and \!" denotes the brand of logical implication formalized by the logic in question In this paper, although essentially subscribing to this view, we propose that the logic to be chosen for this endeavour be a Terminological Logic (TL): accordingly, the IR task becomes that of singling out those documents d such that d q, where d and q are terms of the chosen TL and \ " denotes subsumption between terms We call this the terminological model of IR TLs are particularly suitable for modelling IR; in fact, they can be employed: 1) in representing documents under a variety of aspects (eg structural, layout, semantic content); 2) in representing queries; 3) in representing lexical, \thesaural" knowledge The fact that a single logical language can be used for all these representational endeavours ensures that all these sources of knowledge will participate in the retrieval process in a uniform and principled way In this paper we introduce Mirtl, a TL for modelling IR according to the above guidelines; its syntax, formal semantics and inferential algorithm are described This paper appears in the Proceedings of ACM SIGIR93, 16th International Conference on Research and Development in Information Retrieval, Pittsburgh, PA, 1993, pages 298{307

/pdf/a-model-of-information-retrieval-based-on-a-terminological-4d1n8rdb4d.pdf

A model of information retrieval based on a terminological logic

Research on multimedia information retrieval (MIR) has recently witnessed a booming interest. A prominent feature of this research trend is its simultaneous but independent materialization within several fields of computer science. The resulting richness of paradigms, methods and systems may, on the long run, result in a fragmentation of efforts and slow down progress. The primary goal of this study is to promote an integration of methods and techniques for MIR by contributing a conceptual model that encompasses in a unified and coherent perspective the many efforts that are being produced under the label of MIR. The model offers a retrieval capability that spans two media, text and images, but also several dimensions: form, content and structure. In this way, it reconciles similarity-based methods with semantics-based ones, providing the guidelines for the design of systems that are able to provide a generalized multimedia retrieval service, in which the existing forms of retrieval not only coexist, but can be combined in any desired manner. The model is formulated in terms of a fuzzy description logic, which plays a twofold role: (1) it directly models semantics-based retrieval, and (2) it offers an ideal framework for the integration of the multimedia and multidimensional aspects of retrieval mentioned above. The model also accounts for relevance feedback in both text and image retrieval, integrating known techniques for taking into account user judgments. The implementation of the model is addressed by presenting a decomposition technique that reduces query evaluation to the processing of simpler requests, each of which can be solved by means of widely known methods for text and image retrieval, and semantic processing. A prototype for multidimensional image retrieval is presented that shows this decomposition technique at work in a significant case.

/pdf/a-model-of-multimedia-information-retrieval-2ojjmjlq0g.pdf

A model of multimedia information retrieval

An approach to the document-retrieval problem that aims to increase the efficiency and effectiveness of document-retrieval systems by exploiting the semantic contents of the documents is presented. The document retrieval problem is delineated, and conceptual document modeling basics and requirements are discussed. An experimental system, the Multimedia Office Server (Multos), which implements some of the document-model concepts described, is presented. >

Carlo Meghini

Papers

Deep learning for decentralized parking lot occupancy detection

The Delos digital library reference model : foundations for digital libraries

A model of information retrieval based on a terminological logic

A model of multimedia information retrieval

Conceptual modeling of multimedia documents