In this work, we try to explore new perspectives in order to better structure an online and distance training offer. This approach is done on the basis of a training offer based on the design of an adapted offer, named adaptive learning design. Our choice is to introduce more artificial intelligence which can support this project even if it is necessary to remain on fundamental educational elements such as constructivism. This approach allows us to better understand the issue of algorithmic mediation to facilitate online and distance learning.

Blockchain is one of the core technologies of the present world. Its decentralized architecture has received extensive research attention. Peer-to-peer technology has given freedom and transparency to the user. With the implementation of blockchain inside investment, finance, or trading platforms the problem of security, transparency, and trust can also be addressed. The most suitable consensus protocol can help in fault tolerance. Distribution of blockchain-like hyper ledger promises strong consistency of state and can perform thousands of operations at the same time. Blockchain technology can be helpful in tracing carbon footprints and can be proven helpful in controlling global warming. This paper illustrates what blockchain is, its different consensus mechanisms, its implementation inside an application, its distribution hyper ledger, and how blockchain can be used for tracing carbon footprints.

PLATFORMIZATION OF ONLINE COACHING AS PART OF THE ACQUISITION OF LANGUAGE SKILLS FOR THE VOLTAIRE PROJECT: THE CASE OF THE DAEU AT GUSTAVE EIFFEL UNIVERSITY

This work focuses on the use of the Markov chain to optimize the learning path for learner. We observe that most of them during their online courses can be are victims of boredom which pushes them to give up their courses and consequently do not obtain the certificate, of which they had the ambition to obtain the formation and the diploma. We based our works on Q-Learning methods that seem to meet our need to satisfy students’ demand for titles and certificates without getting bored.

Many businesses have been positively impacted by electronic commerce (ecommerce). It has enabled enterprises and consumers transact business digitally and experience diversity as long as the internet is accessible and there is a gadget to surf the internet. Several governments have gradually adopted electronic payment throughout the country. The Nigerian government has also done a lot of prodding toward the adoption of a cashless economy, which includes embracing ecommerce. As ecommerce expands, so does actual and attempted fraud through this channel. According to the Nigerian Central Bank, electronic fraud reached trillions of Naira by 2021. The purpose of this work was to employ logistic regression as a decision-making tool for detecting fraud in e-commerce platforms at either the virtual or physical point of sale. The main contribution of this research is a model developed using logistic regression for detecting fraud at the point of sale on electronic commerce platforms. The accuracy of the result is 97.8 percent. The result of this study will provide key decision makers in ecommerce firms with information on fraud patterns on their ecommerce platforms, this will enable them take quick actions to forestall these fraudulent attempts. Further research should be carried out using data from other developing countries.

Electronic fraud is a problem that has become a source of concern for businesses of all sizes. Electronic fraud is increasing the margin of loss as criminals go beyond brick and mortar enterprises to target firms with an online presence and electronic payment methods. The purpose of this research paper is to propose a model which decision makers can use to anticipate threats, provide preventive measures and calculate percentage gain in income following execution of the preventive measures. This will assist in safeguarding businesses and consumers from electronic fraud while using selected electronic payment channels. The model will help minimize e-payment fraud and increase customer adoption of electronic payments. The data used was obtained from the Central Bank of Nigeria. The models output offers decision makers in banks and financial technology firms with historical data on the amount of e-payment fraud on each of the selected channels. The model also allows decision makers forecast cyber fraud on various e-payment channels. Similarly, the model offers methods for implementing preventive measures as well as a percentage gain in income following execution. These findings will help to minimize e-payment fraud and increase customer adoption of electronic payments.

Information about the paper titled "ANALYSIS OF LEARNERS' ATTITUDES DURING ONLINE COURSES: DEMONSTRATION OF THE NOTION OF RESILIENCE AND STRUCTURAL APPROACHES ON THE ROLE OF THE PLATFORM FOR THE NOTION OF RESILIENCE" at IATED Digital Library

In this work, we try to show that the contribution of a chatbot in an e-learning platform can bring more efficiency in learning. The observation being that many learners either abandon their learning. They judge that it’s not very effective, because often their questions are not answered adequately. Our emotional chatbot proposal shows that if we take into account the emotional state of the learner, we can offer him an answer adapted to his concerns that appear during their learning processes

Sero-prevalence studies and statistical modelling have been employed over the years for the estimation and monitoring of Hepatitis B <; C development in Nigeria. However, several issues which include wider coverage of the Nigerian population, unified health records for existing and new patients diagnosed with this disease, and access to the dispersed data often collected in isolation by healthcare professionals are yet to be fully addressed. As a result, there is no definite data across the country in policy formulation. The overall aim of this study is to design and implement an observatory system for monitoring hepatitis C development in Nigeria, which will serve as a decision support system and policy-making tool in the healthcare system. The system collects and integrates data, process and display results on the dashboard, that shows information to decision makers for policy formulation.

How can intelligent agents solve a diverse set of tasks in a data-efficient manner? The disentangled representation learning approach posits that such an agent would benefit from separating out (disentangling) the underlying structure of the world into disjoint parts of its representation. However, there is no generally agreed-upon definition of disentangling, not least because it is unclear how to formalise the notion of world structure beyond toy datasets with a known ground truth generative process. Here we propose that a principled solution to characterising disentangled representations can be found by focusing on the transformation properties of the world. In particular, we suggest that those transformations that change only some properties of the underlying world state, while leaving all other properties invariant, are what gives exploitable structure to any kind of data. Similar ideas have already been successfully applied in physics, where the study of symmetry transformations has revolutionised the understanding of the world structure. By connecting symmetry transformations to vector representations using the formalism of group and representation theory we arrive at the first formal definition of disentangled representations. Our new definition is in agreement with many of the current intuitions about disentangling, while also providing principled resolutions to a number of previous points of contention. While this work focuses on formally defining disentangling - as opposed to solving the learning problem - we believe that the shift in perspective to studying data transformations can stimulate the development of better representation learning algorithms.

In order to build agents with a rich understanding of their environment, one key objective is to endow them with a grasp of intuitive physics; an ability to reason about three-dimensional objects, their dynamic interactions, and responses to forces. While some work on this problem has taken the approach of building in components such as ready-made physics engines, other research aims to extract general physical concepts directly from sensory data. In the latter case, one challenge that arises is evaluating the learning system. Research on intuitive physics knowledge in children has long employed a violation of expectations (VOE) method to assess children's mastery of specific physical concepts. We take the novel step of applying this method to artificial learning systems. In addition to introducing the VOE technique, we describe a set of probe datasets inspired by classic test stimuli from developmental psychology. We test a baseline deep learning system on this battery, as well as on a physics learning dataset ("IntPhys") recently posed by another research group. Our results show how the VOE technique may provide a useful tool for tracking physics knowledge in future research.

Psychlab is a simulated psychology laboratory inside the first-person 3D game world of DeepMind Lab (Beattie et al. 2016). Psychlab enables implementations of classical laboratory psychological experiments so that they work with both human and artificial agents. Psychlab has a simple and flexible API that enables users to easily create their own tasks. As examples, we are releasing Psychlab implementations of several classical experimental paradigms including visual search, change detection, random dot motion discrimination, and multiple object tracking. We also contribute a study of the visual psychophysics of a specific state-of-the-art deep reinforcement learning agent: UNREAL (Jaderberg et al. 2016). This study leads to the surprising conclusion that UNREAL learns more quickly about larger target stimuli than it does about smaller stimuli. In turn, this insight motivates a specific improvement in the form of a simple model of foveal vision that turns out to significantly boost UNREAL's performance, both on Psychlab tasks, and on standard DeepMind Lab tasks. By open-sourcing Psychlab we hope to facilitate a range of future such studies that simultaneously advance deep reinforcement learning and improve its links with cognitive science.

Amos DAVID, Universite de Lorraine, Nancy, France ; African University of Science and Technology, Abuja, Nigeria; Laboratory DICEN-IDF, Paris, France Nadine NDJOCK, ESSTIC/Universite de Yaounde 2-Soa ; Laboratory YMIS, Yaounde, Cameroun Sub-theme : Foundations and methods for KO Title : Big Data, Knowledge Organization and Decision Making – Opportunities and limit Keywords: Big data, knowledge organization, Decision-making, Information visualization, Data analysis Objectives A concept that is currently attracting much interest in the field of information science is the concept of Big Data. How does it relate to Knowledge Organization and to decision making? What are its opportunities and limits? These are some questions we try to answer in this paper. This paper has been mainly inspired by Challenges and Opportunities with Big Data: A white paper prepared for the Computing Community Consortium committee of the Computing Research Association (Agrawal, et al., 2012) and the theme of the biennial conference “Transition from Observation to Knowledge to Intelligence” started in 2014 and to be organized now by the newly created ISKO-West Africa chapter (DAVID, A., & UWADIA, C. 2016) The Big Data pipeline is presented as below (Agrawal, et al., 2012) Summarizing the challenges with Big Data, (Agrawal, et al., 2012) states that “Heterogeneity, scale, timeliness, complexity, and privacy problems with Big Data impede progress at all phases of the pipeline that can create value from data. The problems start right away during data acquisition, when the data tsunami requires us to make decisions, currently in an ad hoc manner, about what data to keep and what to discard, and how to store what we keep reliably with the right metadata. Much data today is not natively in structured format; for example, tweets and blogs are weakly structured pieces of text, while images and video are structured for storage and display, but not for semantic content and search: transforming such content into a structured format for later analysis is a major challenge. Data analysis is a clear bottleneck in many applications, both due to lack of scalability of the underlying algorithms and due to the complexity of the data that needs to be analyzed. Finally, presentation of the results and its interpretation by non-technical domain experts is crucial to extracting actionable knowledge”. Methods Within the framework of our study using the competitive intelligence approach to problem decision solving, we have also identified the problem associated with the use of normal watch technique which is closely related to the current approach employed when referring to the potentials of Big Data (DAVID, A., 2016) In the watch approach to problem solving, information is first collected, then verified for validity and then examine the possible application for solving the problem at hand. We believe that this approach is generally not appropriate since it produces a lot of noise – gathering information that will be discarded because not applicable to the problem at hand. CVI model for information use in decision making process Based on the observation above, we propose that the steps be inverted, starting with the identification of the possible use of the information for solving the problem at hand, verify the relevance and validity of the information sources and then collect the relevant information IVC model for information use in decision making process To facilitate understanding of results, we propose to extend the information system by integrating visualization tools to present the information. It is from these visual presentations, also based on the indicators, which in turn are derived from the basic information, which allow a better interpretation of all the information collected (Ndjock, 2017). Main results Our proposals in terms of models and tools have been implemented in some systems. Added-value information through visualization has been applied in two applications, which will be developed in the full paper: The first application was developed for the PhD thesis of Dr. Nadine NDJOCK (Ndjock, 2017), which concerns optimization of decision-making process by the visualization of information through the concept of observatory, applied to the educational system in Cameroon. The decision-maker obtains the evolution of indicators, which enables to guide a strategic decision such as the efficient management of personnel or adjustment of the training program. The second application concerns - ISKO (International Society for Knowledge Organization) membership management system. The system is used to manage the members of the association. Not only can Executive Office and Chapter Administrators create, access and modify member profiles, they can obtain added-value information through visualization tools that allow deployment of development strategies. Conclusion Technologies have been proposed for managing Bid Data, such as Hadoop file system. Some layers on Hadoop have been proposed to enhance access and information systems integrating query and visualization techniques. However, a lot still need to be done to conceptualize the method to use before information collection in view of Big Data management for analysis and for decision making. We are currently working on the concept of observatory systems. References Agrawal, D., Bernstein, P., Elisa, B., Susan, D., Umeshwar, D., Franklin, M., . . . Papakonstantinou, Y. (2012). Challenges and Opportunities with Big Data: A white paper prepared for the Computing Community Consortium. DAVID, A. (2016). From data to intelligence - Strategic decision making through information system Transition from Observation to Knowledge to Intelligence. Lagos, Nigeria: ISBN 978-2-9546760. DAVID, A., & UWADIA, C. (2016). Transition from Observation to Knowledge to Intelligence . Lagos, Nigeria: ISBN 978-2-9546760-3-6. Duhigg , C. (2012). The Power of Habit: Why we do what we do in life and business. Doubleday Canada, Random House Canada Ltd , pp. 190. Ndjock, F. N. (2017). From observation to decision-making : How an information system can improve strategic decision-making. International Journal of Social Science and Technology. Vol. 2 No. 3 , pp 89-99.

We introduce a new dataset of logical entailments for the purpose of measuring models' ability to capture and exploit the structure of logical expressions against an entailment prediction task. We use this task to compare a series of architectures which are ubiquitous in the sequence-processing literature, in addition to a new model class—PossibleWorldNets—which computes entailment as a "convolution over possible worlds". Results show that convolutional networks present the wrong inductive bias for this class of problems relative to LSTM RNNs, tree-structured neural networks outperform LSTM RNNs due to their enhanced ability to exploit the syntax of logic, and PossibleWorldNets outperform all benchmarks.

Animals execute goal-directed behaviours despite the limited range and scope of their sensors. To cope, they explore environments and store memories maintaining estimates of important information that is not presently available. Recently, progress has been made with artificial intelligence (AI) agents that learn to perform tasks from sensory input, even at a human level, by merging reinforcement learning (RL) algorithms with deep neural networks, and the excitement surrounding these results has led to the pursuit of related ideas as explanations of non-human animal learning. However, we demonstrate that contemporary RL algorithms struggle to solve simple tasks when enough information is concealed from the sensors of the agent, a property called "partial observability". An obvious requirement for handling partially observed tasks is access to extensive memory, but we show memory is not enough; it is critical that the right information be stored in the right format. We develop a model, the Memory, RL, and Inference Network (MERLIN), in which memory formation is guided by a process of predictive modeling. MERLIN facilitates the solution of tasks in 3D virtual reality environments for which partial observability is severe and memories must be maintained over long durations. Our model demonstrates a single learning agent architecture that can solve canonical behavioural tasks in psychology and neurobiology without strong simplifying assumptions about the dimensionality of sensory input or the duration of experiences.

We consider the general problem of modeling temporal data with long-range dependencies, wherein new observations are fully or partially predictable based on temporally-distant, past observations. A sufficiently powerful temporal model should separate predictable elements of the sequence from unpredictable elements, express uncertainty about those unpredictable elements, and rapidly identify novel elements that may help to predict the future. To create such models, we introduce Generative Temporal Models augmented with external memory systems. They are developed within the variational inference framework, which provides both a practical training methodology and methods to gain insight into the models' operation. We show, on a range of problems with sparse, long-term temporal dependencies, that these models store information from early in a sequence, and reuse this stored information efficiently. This allows them to perform substantially better than existing models based on well-known recurrent neural networks, like LSTMs.

This paper adopts the concepts of observatory and competitive intelligence (CI) to model a system that will generate better insights for decisionmakers in the solid waste industry. The first part of this work is to design and develop a data warehouse (DWH) of solid waste statistics using data assembled from disparate sources. Our methodology of design is the entity relationship diagram (ERD) and our implementation tool is MySQL running on phpMyAdmin. The second part of our work will be to turn our developed DWH into a Web application using the Yii PHP component framework. Our findings indicated that the application of both concepts of observatory and CI lead to better insights for decision-makers and hence better organizational performance.

The amount of information in term of documents, available to users as a result of information retrieval process for the purpose of resolution of decision problems is a major factor that determines whether economically viable decisions would be made or not.Various works in the literature had addressed the challenges of representing the documents with key terms (generated from the document) as well as the variations in the meaning of each key terms.In this work, a document representation scheme that is based on the key terms generated from the documents and their usage was developed.To realize this document representation scheme, a computational model for capturing document usage was designed with the use of attribute value pair technique of document annotation.The document usage model designed was applied in the development of a Competitive Intelligence based Document Usage Creation and Exploration system that is currently under development.A preliminary evaluation of the document usage model based on cosine similarity function between user query and documents set was carried out.The result obtained shows that representing documents in terms of their usage can enhance the quality of information search results as documents that would hitherto be considered not relevant to user query are found to be ranked very relevant based on previous usages.