Youcef Ouinten

OLAP textual aggregation approach using the Google similarity distance

Data warehousing and online analytical processing OLAP are essential elements to decision support. In the case of textual data, decision support requires new tools, mainly textual aggregation functions, for better and faster high level analysis and decision making. Such tools will provide textual measures to users who wish to analyse documents online. In this paper, we propose a new aggregation function for textual data in an OLAP context based on the K-means method. This approach will highlight aggregates semantically richer than those provided by classical OLAP operators. The distance used in K-means is replaced by the Google similarity distance which takes into account the semantic similarity of keywords for their aggregation. The performance of our approach is analysed and compared to other methods such as Topkeywords, TOPIC, TuBE and BienCube. The experimental study shows that our approach achieves better performances in terms of recall, precision, F-measure complexity and runtime.

Simulation of Mobile Ad hoc Routing Strategies

Several routing protocols for mobile ad hoc networks have been proposed in the literature, each one based on a different strategic: proactive, reactive and hybrid. Some of these protocols have been studied and their performance is evaluated in detail. However a comparison between them is lacking to help determine an optimal strategy. This paper addresses this issue by comparing three protocols DSDV, AODV and CBRP from different MANET routing strategies. Performance is evaluated with respect to packets delivery ratio, end-to-end delay and normalized routing load for a given mobility mode, traffic load and network size by using Ns2 simulator.

Vertical fragmentation of data warehouses using the FP-Max algorithm

Vertical partitioning is a technique used to reduce disk access, when executing a given set of queries, by minimizing the access to irrelevant instance variables. In this paper we use the FP-Max data mining algorithm, for extracting frequent item set attributes. The frequently accessed instance variables are, then, grouped as vertical class fragments. We study the application of this approach to sets of queries on large databases and data warehouses. We used two benchmarks with various minimum support levels and we compare our results with the results of the approach using the Apriori data mining technique. The partitioning solution obtained produces an improvement of 14% for large data bases and 19% for data warehouse compared to a solution without partitioning.

GOTA - Using the Google Similarity Distance for OLAP Textual Aggregation

With the tremendous growth of unstructured data in the Business Intelligence, there is a need for incorporating textual data into data warehouses, to provide an appropriate multidimensional analysis (OLAP) and develop new approaches that take into account the textual content of data. This will provide textual measures to users who wish to analyse documents online. In this paper, we propose a new aggregation function for textual data in an OLAP context. For aggregating keywords, our contribution is to use a data mining technique, such as kmeans, but with a distance based on the Google similarity distance. Thus our approach considers the semantic similarity of keywords for their aggregation. The performance of our approach is analyzed and compared to another method using the k-bisecting clustering algorithm and based on the Jensen-Shannon divergence for the probability distributions. The experimental study shows that our approach achieves better performances in terms of recall, precision,F-measure complexity and runtime.

A fast token based algorithm for multiple resources sharing in distributed systems

This paper presents a new algorithm for simultaneous resources sharing by several processes in distributed system. This algorithm treats the problem of message complexity which is an important factor in such a problem, and the waiting time of the requesting sites which can be considered as a quality measure of the distributed algorithms. In this algorithm, k tokens are used to resolve the problem with a sensible method of tokens exchanging to minimize the number of exchanged messages for every entry to the CS. The algorithm uses also a fast protocol which minimizes the waiting time between the launch of the requests and the access to the CS.

Efficiently mining frequent itemsets applied for textual aggregation

Text mining approaches are commonly used to discover relevant information and relationships in huge amounts of text data. The term data mining refers to methods for analyzing data with the objective of finding patterns that aggregate the main properties of the data. The merger between the data mining approaches and on-line analytical processing (OLAP) tools allows us to refine techniques used in textual aggregation. In this paper, we propose a novel aggregation function for textual data based on the discovery of frequent closed patterns in a generated documents/keywords matrix. Our contribution aims at using a data mining technique, mainly a closed pattern mining algorithm, to aggregate keywords. An experimental study on a real corpus of more than 700 scientific papers collected on Microsoft Academic Search shows that the proposed algorithm largely outperforms four state-of-the-art textual aggregation methods in terms of recall, precision, F-measure and runtime.

Textual aggregation approaches in OLAP context: A survey

Abstract In the last decade, OnLine Analytical Processing (OLAP) has taken an increasingly important role as a research field. Solutions, techniques and tools have been provided for both databases and data warehouses to focus mainly on numerical data. however these solutions are not suitable for textual data. Therefore recently, there has been a huge need for new tools and approaches that treat and manipulate textual data and aggregate it as well. Textual aggregation techniques emerge as a key tool to perform textual data analysis in OLAP for decision support systems. This paper aims at providing a structured and comprehensive overview of the literature in the field of OLAP Textual Aggregation. We provide a new classification framework in which the existing textual aggregation approaches are grouped into two main classes, namely approaches based on cube structure and approaches based on text mining. We discuss and synthesize also the potential of textual similarity metrics, and we provide a recent classification of them.

A New Tool for Textual Aggregation In Information Retrieval

We present in this paper a system for textual aggregation from scientific documents in the online analytical processing (OLAP) context. The system extracts keywords automatically from a set of documents according to the lists compiled in the Microsoft Academia Search web site. It gives the user the possibility to choose their methods of aggregation among the implemented ones. That is TOP-Keywords, TOPIC, TUBE, TAG, BienCube and GOTA. The performance of the chosen methods, in terms of recall, precision, F-measure and runtime, is investigated with two real corpora ITINNOVATION and OHSUMED with 600 and 13,000 scientific articles respectively, other corpora can be integrated to the system by users.

Improving Index Selection Accuracy for Star Join Queries Processing: An Association Rules Based Approach

Nowadays, the new technologies for Business Intelligence as DataWarehouse, OLAP, Data Mining, emerged and are needed for the managerial process. In the area of decision support systems, a basic role is held by a data warehouse which is an online repository for decision support applications using complex star join queries. Answering such queries efficiently is often difficult due to the complex nature of both the data and the queries. One of the most challenging tasks for the data warhouse administrator (DWA) is the selection of a set of indexes to attain optimal performance for a given workload under storage constraint. The problem is shown to be NP-hard since it involves searching a vast space of possible configurations. It is very much important to extract meaningful information from the workload which represents the major step towards building relevant indexes. This paper presents an approach for selecting an optimized index configuration using association rules with Apriori algorithm which can drive to understand with more accuracy the attributes correlation. This helps to recommend an index set that closely match the requirements of the provided workload. Experimented using the ABP-1 benchmark, our proposed approach achieves good performance compared with previous studies.

An improved approach for automatic selection of multi-tables indexes in ralational data warehouses using maximal frequent itemsets

System performance for data warehouses is crucially dependent on its physical design in which one of the most challenging tasks is the selection of an appropriate set of indexes for a representative workload under storage constraint. The problem becomes even more complex for multi-tables indexes such as bitmap join indexes, since it involves searching a vast space of possible configurations. Queries references to attributes and their frequencies play an important role in determining the efficiency of the selected indexes. In this paper, we consider the index selection as a typical frequent itemsets mining problem. The indexes are built with combinations of attributes, viewed as items. The queries in the workload, viewed as transactions, are described by the attributes they involve. The foundation of our approach is the concept of maximal frequent itemsets. This data mining technique helps to discover strong correlations among attributes such that the presence of some attributes in a query will imply the presence of some other attributes. Moreover, by avoiding the generation of redundent indexes, the proposed approach leads to a solution that expresses the set of relevant indexes in a more succinct way. Consequently, it guarantees the reduction of the storage space requirements. Unlike previous approaches that focus on the configuration leading to the minimum workload cost, we suggest to consider a set of optimized solutions and we propose a metric for measuring profit-effectiveness that helps to pick up the most promising one. Through a set of experiments on the ABP-1 benchmark, we show that our approach achieves better performance compared to similar methods, with significant savings in index storage.