Alberto Córdoba

Pattern Matching Techniques to Identify Syntactic Variations of Tags in Folksonomies

Folksonomies offer an easy method to organize information in the current Web. This fact and their collaborative features have derived in an extensive involvement in many Social Web projects. However they present important drawbacks regarding their limited exploring and searching capabilities, in contrast with other methods as taxonomies, thesauruses and ontologies. One of these drawbacks is an effect of its flexibility for tagging, producing frequently multiple syntactic variations of a same tag. In this paper we study the application of two classical pattern matching techniques, Levenshtein distance for the imperfect string matching and Hamming distance for the perfect string matching, to identify syntactic variations of tags.

A distributed data gathering algorithm for wireless sensor networks with uniform architecture

Wireless sensor networks (WSNs) are commonly used for continuously monitoring applications. Data gathering and in-network processing are techniques to efficiently gather the information from sensor nodes. Usually, these techniques have been analyzed taking into account a global knowledge of the system in order to select the best path to gather the information or to establish a routing function. In this paper we propose a novel distributed data gathering algorithm for wireless networks and evaluate the effect of having a different number of cluster heads transmitting to the base station. We focus the attention on uniformly distributed nodes in a given area to evaluate the effects on the energy consumption. Our proposal distributes uniformly the energy consumption between the complete set of sensors in the system resulting in a similar longevity for all sensors in the system

Efficient leader election in complete networks

Leader election is a fundamental problem in distributed computing and it has been studied assuming various computation models and network topologies. This paper analyzes the algorithms for leader election in complete networks using asynchronous communication channels. We present a novel algorithm that reduces the information necessary to select a leader compared with other leader election algorithms for complete networks. In this paper, the algorithm works without sense of direction. And, it does not require to know the number of nodes in the system. Our proposal requires O(n) messages and O(n) time, where n is the number of nodes in the system, to elect a leader.

Ontology Based Road Traffic Management

Every time an abnormal situation happens on the road, danger situations are found and they need the skills of the drivers in order to cope with them. In this paper, we propose to integrate an ontology inside each vehicle to provide them with reasoning capabilities and avoid the use of a central decision point. Our system benefits traffic management in emergency situations with immediate decisions where each vehicle is a decision point, reacts considering its neighbouring vehicles, and collaborates with them to reach consensus in real time.

A Distributed Data Gathering Algorithm for Wireless Sensor Networks with Uniform Architecture

Wireless sensor networks (WSNs) are commonly used for continuously monitoring applications. Data gathering and in-network processing are techniques to efficiently gather the information from sensor nodes. Usually, these techniques have been analyzed taking into account a global knowledge of the system in order to select the best path to gather the information or to establish a routing function. In this paper we propose a novel distributed data gathering algorithm for wireless networks and evaluate the effect of having a different number of cluster heads transmitting to the base station. We focus the attention on uniformly distributed nodes in a given area to evaluate the effects on the energy consumption. Our proposal distributes uniformly the energy consumption between the complete set of sensors in the system resulting in a similar longevity for all sensors in the system

An O(n) distributed deadlock resolution algorithm

This paper shows a new distributed algorithm for deadlock detection and resolution under the single-resource request model that highly improves the complexity measurements of previous proposals. The algorithm has a communication cost of 2n - 1 messages and a latency of n /spl middot/ T for a deadlock cycle of n processes, where T is the inter-site communication delay. The algorithm achieves this improvement even verifying the strongest correctness criteria considered in previous works: it resolves all deadlocks in finite time and does not resolve false deadlocks.

A safe algorithm for resolving OR deadlocks

Deadlocks in the OR model are usually resolved by aborting a deadlocked process. Prior algorithms for the same model sometimes abort nodes needlessly wasting computing resources. This paper presents a new deadlock resolution algorithm for the OR model that satisfies the following correctness criteria: (Safety) the algorithm does not resolve false deadlocks; (Liveness) the algorithm resolves all deadlocks in finite time. The communication cost of the algorithm is similar to that of previous nonsafe proposals. The theoretical cost has been validated by simulation. In addition, different algorithm initiation alternatives have been analyzed in order to reduce the latency of deadlocks.

AModified O(n) Leader Election Algorithm for Complete Networks

This paper presents a modified leader election algorithm for complete networks without sense of direction. The original algorithm, introduced by Villadangos et al. in (2005), had the aim of reducing the number of exchanged messages in order to select a leader. However, the original O(n) algorithm fails to choose a leader on several occasions. A modified algorithm, which eliminates the problems that cause the wrong behaviour, is proposed. It is formally proved that the new algorithm verifies the correctness criteria that consist of selecting a unique leader in every case. Additionally, the algorithm maintains the O(n) complexity in both messages and time, where n is the number of nodes in the system

A low communication cost algorithm for distributed deadlock detection and resolution

This work introduces a new distributed history-based algorithm for deadlock detection and resolution under the single-unit request model. The algorithm has a communication cost of O(n /spl middot/ log n) messages for a deadlock cycle of n processes. This low cost is achieved by means of two mechanisms. On one hand, to reduce the number of instance initializations, a node compares the priorities between its immediate successor and all its predecessors, starting the algorithm only if an antagonistic conflict is given. On the other hand, each instance of the algorithm runs at a certain detection level and does not retransmit probes created by instances running at a lower detection level. An instance can switch to a higher level depending on the information received in the probes.

A Tag Clustering Method to Deal with Syntactic Variations on Collaborative Social Networks

Folksonomies have emerged as a common way of annotating and categorizing content using a set of tags that are created and managed in a collaborative way. Tags carry the semantic information within a folksonomy, and provide thus the link to ontologies. The appeal of folksonomies comes from the fact that they require a low effort for creation and maintenance since they are community-generated. However they present important drawbacks regarding their limited navigation and searching capabilities, in contrast with other methods as taxonomies, thesauruses and ontologies. One of these drawbacks is an effect of its flexibility for tagging, producing frequently multiple syntactic variations of a same tag. Similarity measures allow the correct identification of tag variations when tag lengths are greater than five symbols. In this paper we propose the use of cosine relatedness measures in order to cluster tags with lengths lower or equal than five symbols. We build a discriminator based on the combination of a fuzzy similarity and a cosine measures and we analyze the results obtained.