Saleh Alshomrani

Image de-noising using noise ratio estimation, K-means clustering and non-local means-based estimator

One of the key issues in removing random-valued impulse noise from digital images using switching filters is the impulse noise detection. Impulse noise is a random, spiked variation in the brightness of the image. In this paper, a new impulse noise detection algorithm is presented that is based on Noise ratio Estimation and a combination of K-means clustering and Non-Local Means based filter (NEK-NLM). Luo-statistic is employed as a non-local means based estimator. The novelty of this work lies in the introduction of a pre-processing step of noise ratio estimation before noise detection, this estimation allows us to select suitable parameters for the noise detection algorithm. In noise filtering stage, nonlocal-means estimator is applied for restoring noisy pixels to their actual values. Using real world datasets, this paper shows that the impulse noise can be removed effectively. Extensive comparison of simulation results with the already published results show that the proposed method outperforms most of the existing impulse noise removal techniques both in terms of noise detection and image restoration.

2LPA-RTDW: A Two-Level data Partitioning Approach for Real-time Data Warehouse

In order to explore the most recent data and react faster to changes of business conditions, organizations consider Real-Time Data Warehousing (RTDW) as a powerful technique to achieve OLAP (On Line Analytical Processing) analyses and business intelligence (BI). OLAP analyses are complex since they query several relational tables with huge volumes. In order to deal with this volumetry, several optimization techniques have been proposed in the literature as materialized views and data partitioning. Partitioning is an effective method to increase query efficiency in a data warehouse. This paper proposes a novel data partitioning approach for real-time data warehouse, called 2LPA-RTDW (Two-Level data Partitioning Approach for Real-Time Data Warehouse) by allowing unbalance of data amount in each partition while taking into account user requirements. We have evaluated the proposed approach using the new TPC-DS1 benchmark; the preliminary results show that the approach is quite interesting.

The Power of a Model-Driven Approach to Handle Evolving Data Warehouse Requirements.

The Data Warehouse (DW) is characterized by complex architecture, specific modeling and design approaches. It integrates data issued from operational data sources in order to meet decision-makers’ needs by providing answers for OLAP queries (On-Line Analytical Processing). In practice, both data source models and decision-makers’ analytical requirements evolve over time and, therefore, lead to changes in the DW multidimensional model. In this evolving context, we have developed the DWE (Data Warehouse Evolution) framework. DWE automatically propagates the changes of the data source data-model on the DW data-model. This paper proposes a model-driven approach for extending DWE in order to consider a further related evolutionary aspect: The evolution of decision-makers’ needs. It deals with the propagation of these evolutions on the DW multidimensional model. This approach relies on a classification of evolution scenarios and a set of transformation rules for the identification of evolut ion operations to apply on the DW.

Decision Tree Transformation for Knowledge Warehousing

Explicit knowledge extracted from data, formalized tacit knowledge from experts or even knowledge existing in business sources may be in several heterogeneous formal representations and structures: as rules, models, functions, etc. However, a knowledge warehouse should solve this structural heterogeneity before storing knowledge. This requires specific tasks of harmonizing. This paper first presents our proposed definition and architecture of a knowledge warehouse, and then presents some languages for knowledge representations as particular the MOT (Modeling with Object Types) language. In addition, we suggest a metamodel for the MOT, and a metamodel for the explicit knowledge obtained using decision trees technique. As we aim to represent knowledge having different modeling formalisms into MOT, as a unified model, then we suggest a set of transformation rules that assure the move from the decision tree source model into the MOT target model. This work is still in progress, it is currently completed with tranformations for additional.

CNC-NOS: Class noise cleaning by ensemble filtering and noise scoring

Abstract Obtaining data in the real world is subject to imperfections and the appearance of noise is a common consequence of such flaws. In classification, class noise will deteriorate the performance of a classifier, as it may severely mislead the model building. Among the strategies emerged to deal with class noise, the most popular is that of filtering. However, instance filtering can be harmful as it may eliminate more examples than necessary or produce loss of information. An ideal option would be relabeling the noisy instances, avoiding losing data, but instance correcting is harder to achieve and may lead to wrong information being introduced in the dataset. For this reason, we advance a new proposal based on an ensemble of noise filters with the goal not only of accurately filtering the mislabeled instances, but also correcting them when possible. A noise score is also applied to support the filtering and relabeling process. The proposal, named CNC-NOS (Class Noise Cleaner with Noise Scoring), is compared against state-of-the-art noise filters and correctors, showing that it is able to deliver a quality training instance set that overcomes the limitations of such techniques, both in terms of classification accuracy and properly treated instances.

CobWeb multidimensional model: From OLAP to tag cloud

The development of Internet is permanently increasing the number of documents and the volumes of data available and exchanged through the Web. This documentary information constitutes an interesting source for the decision-making analysis. Therefore, it is essential to provide decision-makers with efficient tools to analyze the textual data enclosed in documents. In this paper, first we present a generic multidimensional model called “CobWeb”, which extends the multidimensional galaxy. CobWeb is dedicated to the OLAP (OnLine Analytical Processing) of XML documents, and based on the concept of facet. It aims to provide decision-makers with facilities in expressing their analytical queries along with an appropriate vision of the data. Secondly, we introduce our software prototype called MQF (Multidimensional Query based on Facets) for querying the CobWeb multidimensional document warehouse. Query results are displayed as multidimensional tables, and as a cloud of tags.