Informatique

We propose an automatic classification system of movie genres based on different features from their textual synopsis. Our system is first trained on thousands of movie synopsis from online open databases, by learning relationships between textual signatures and movie genres. Then it is tested on other movie synopsis, and its results are compared to the true genres obtained from the Wikipedia and the Open Movie Database (OMDB) databases. The results show that our algorithm achieves a classification accuracy exceeding 75%.

Data in real world applications are in most cases linguistic information that are ambiguous and uncertain. Hence, such data should be handled by fuzzy set representation schemes to increase expressiveness and comprehensiveness. Moreover, mining these data requires ways to generate automatically useful information/knowledge through a set of fuzzy rules. This paper proposes a novel system called FuAGGE that stands for Fuzzy Automatic Generator Genetic Expression. The FuAGGE approach uses a grammar based   evolutionary technique. The grammar is expressed in the Backus Naur Form (BNF) and represents a fuzzy set covering method. The grammar is mapped into programs that are themselves implementations of fuzzy rule-based learners. Binary strings are used as inputs to the mapper along with the BNF grammar. These binary strings represent possible potential solutions resulting from the initializer component and the building blocks from Weka, a workbench that contains a collection of visualiza...

We propose an automatic classification system of movie genres based on different features from their textual synopsis. Our system is first trained on thousands of movie synopsis from online open databases, by learning relationships between textual signatures and movie genres. Then it is tested on other movie synopsis, and its results are compared to the true genres obtained from the Wikipedia and the Open Movie Database (OMDB) databases. The results show that our algorithm achieves a classification accuracy exceeding 75%.

A lot of hype has recently been generated around deep learning, a group of artificial intelligence approaches able to break accuracy records in pattern recognition. Over the course of just a few years, deep learning revolutionized several research fields such as bioinformatics or medicine. Yet such a surge of tools and knowledge is still in its infancy in ecology despite the ever-growing size and the complexity of ecological datasets. Here we performed a literature review of deep learning implementations in ecology to identify its benefits in most ecological disciplines, even in applied ecology, up to decision makers and conservationists alike. We also provide guidelines on useful resources and recommendations for ecologists to start adding deep learning to their toolkit. At a time when automatic monitoring of populations and ecosystems generates a vast amount of data that cannot be processed by humans anymore, deep learning could become a necessity in ecology.

This paper presents a new steganalysis method that uses a deep belief network (DBN) as a classifier for audio files. It has been tested on three steganographic techniques: StegHide, Hide4PGP and FreqSteg. The results were compared to two other existing robust steganalysis methods based on support vector machines (SVMs) and Gaussian mixture models (GMMs). Afterwards, another classification task aiming at identifying the type of steganographic applied or not to the speech signal was carried out. The results of this four-way classification show that in most cases, the proposed DBN-based steganalysis method gives higher classification rates than the two other steganalysis methods based on SVMs and GMMs.

Methods in Ecology and Evolution This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

Background: Many apps have been designed to remotely assess clinical status and monitor symptom evolution in persons with multiple sclerosis (MS). These may one day serve as an adjunct for in-person assessment of persons with MS, providing valuable insight into the disease course that is not well captured by cross-sectional snapshots obtained from clinic visits. Objective: This study sought to review the current literature surrounding apps used for remote monitoring of persons with MS. Methods: A scoping review of published articles was conducted to identify and evaluate the literature published regarding the use of apps for monitoring of persons with MS. PubMed/Medline, EMBASE, CINAHL, and Cochrane databases were searched from inception to January 2022. Cohort studies, feasibility studies, and randomized controlled trials were included in this review. All pediatric studies, single case studies, poster presentations, opinion pieces, and commentaries were excluded. Studies were assessed for risk of bias using the Scottish Intercollegiate Guidelines Network, when applicable. Key findings were grouped in categories (convergence to neurological exam, feasibility of implementation, impact of weather, and practice effect), and trends are presented. In a parallel systematic search, the Canadian Apple App Store and Google Play Store were searched to identify relevant apps that are available but have yet to be formally studied and published in peer-reviewed publications. Results: We included 18 articles and 18 apps. Although many MS-related apps exist, only 10 apps had published literature supporting their use. Convergence between app-based testing and the neurological exam was examined in 12 articles. Most app-based tests focused on physical disability and cognition, although other domains such as ambulation, balance, visual acuity, and fatigue were also evaluated. Overall, correlations between the app versions of standardized tests and their traditional counterparts were moderate to strong. Some novel app-based tests had a stronger correlation with clinician-derived outcomes than traditional testing. App-based testing correlated well with the Multiple Sclerosis Functional Composite but less so with the Expanded Disability Status Scale; the latter correlated to a greater extent with patient quality of life questionnaire scores. Conclusions: Although limited by a small number of included studies and study heterogeneity, the findings of this study suggest that app-based testing demonstrates adequate convergence to traditional in-person assessment and may be used as an adjunct to and perhaps in lieu of specific neurological exam metrics documented at clinic visits, particularly if the latter is not readily accessible for persons with MS.

A lot of hype has recently been generated around deep learning, a group of artificial intelligence approaches able to break accuracy records in pattern recognition. Over the course of just a few years, deep learning revolutionized several research fields such as bioinformatics or medicine. Yet such a surge of tools and knowledge is still in its infancy in ecology despite the ever-growing size and the complexity of ecological datasets. Here we performed a literature review of deep learning implementations in ecology to identify its benefits in most ecological disciplines, even in applied ecology, up to decision makers and conservationists alike. We also provide guidelines on useful resources and recommendations for ecologists to start adding deep learning to their toolkit. At a time when automatic monitoring of populations and ecosystems generates a vast amount of data that cannot be processed by humans anymore, deep learning could become a necessity in ecology.