Time series forecasting has become a very intensive field of research, which is even increasing in recent years. Deep neural networks have proved to be powerful and are achieving high accuracy many application fields. For these reasons, they one the most widely used methods machine learning solve problems dealing with big data nowadays. In this work, time problem initially formulated along its mathematical fundamentals. Then, common deep architectures that currently being successfully...

The economic sector is one of the most important pillars countries. Economic activities industry are intimately linked with ability to meet their needs for electricity. Therefore, electricity forecasting a very task. It allows better planning and management energy resources. Several methods have been proposed forecast consumption. In this work, predict monthly consumption sector, we develop novel approach based on ensemble learning. Our combines three models that proved successful in field,...

Views materialization is a powerful technique for optimization of query evaluation. However, when data sources are updated, these materialized views (MVs) should be maintained in order to get appropriate queries’ answers. This work provides an overview the research field view maintenance, focusing on maintenance issues and dimensions that cover most important ones proposed literature review. We provide classification review state-of-the-art approaches according several parameters...

Traditional data warehouses have played a key role in decision support system until the recent past. However, rapid growing of generation by current applications requires new warehousing systems: volume and format collected datasets, source variety, integration unstructured powerful analytical processing. In age Big Data, it is important to follow this pace adapt existing warehouse systems overcome issues challenges. paper, we focus on over big data. We discuss limitations traditional ones....

Background: A study about healthcare resources can improve decisions regarding the allotment and mobilization of medical to better guide future investment in health sector.Aim: The aim this work was design implement a decision support system allocation Bejaia region.Methods: To achieve retrospective cohort study, we integrated existing clinical databases from different department sector institutions (an Algerian department) collect information patients January 2015 through December 2015.Data...

Non-technical losses (NTL), especially fraud detection is very important for electricity distribution enterprises. Fraud allows maximizing the effective economic return such This paper provides an approach based on robust exponential and Holt-Winters Smoothing methods. The proposed a procedure that aims to discover fraudulent behavior of consumers goes through three crucial steps: (1) prediction monthly consumption, (2) abnormal consumption electrical meters, (3) cases customers. model was...

The purpose of this review is to explore how the use big data technology improves performance medical warehousing. Indeed, traditional warehousing tools can no longer be used handle volume, variety, and velocity today's data-centric applications. Moreover, Big technologies process streams data, they increase by using a cluster existing networked nodes through powerful processing these streams. In paper, we provide an overview state-of-the-art research issues in field opportunities presented...

An important issue for the electricity distribution companies is non-technical loss (NTL), also known as fraud. This has a significant impact on economies of all countries in world. In this context, we studied problem imbalance between electrical energy invoiced and supplied within Algerian economic sector. article presents an approach to detecting fraud using combination Long Short-Term Memory (LSTM) robust Exponential Holt-Winters Smoothing (EHWS) methods order enhance accuracy efficacy...

To improve query performance materialised views were largely used for such purpose. So far, this technique has been a success as testified by the large and increasing interest among industrial research communities in use of technique. In paper, we investigate related issues most important solutions framework literature review. Then, explore information technology impact on evolution view applications, environments data models. Finally, summarise opportunities future challenges caused...

To improve query performance materialised views were largely used for such purpose. So far, this technique has been a success as testified by the large and increasing interest among industrial research communities in use of technique. In paper, we investigate related issues most important solutions framework literature review. Then, explore information technology impact on evolution view applications, environments data models. Finally, summarise opportunities future challenges caused...

In a highly dynamic environment like the Internet of Things, data is being continuously generated by IoT devices. These amounts make its processing and querying challenging tasks. Materialized views are ideal for query optimization caching results queries. However, one main issues materialization their inconsistency with sources. For this reason, we present novel approach answering management in systems. Our based on new maintenance strategies. This last was carried out factoring treatments...

Background: A modern tele-consultation system improves patients’ monitoring and favors remote assistance in terms of facilitating the daily life to patients. The aim this work is design implement a consultation system. Methods: To achieve work, we identify actors who interact with be developed use cases relating each actor. class model designed derive relational corresponding database. During implementation, have used PHP language MySQL database Results: framework implemented. users...

Speeding up queries processing is an important issue in database management. Materialized views are largely used to address this issue. They have been proven successful for query performance optimization. However, updating data sources of the corresponding view requires maintaining related views. Therefore, a maintenance strategy required. This paper presents novel approach materialized that overcomes limitations prior approaches using parallel “Divide and Conquer” strategy. We modeled...