Abstract
A machine learning approach that is capable of treating data streams presents new challenges and enables the analysis of a variety of real problems in which concepts change over time. In this scenario, the ability to identify novel concepts as well as to deal with concept drift axe two important attributes. This paper presents a technique based on the k-means clustering algorithm aimed at considering those two situations in a single learning strategy. Experimental results performed with data from various domains provide insight into how clustering algorithms can be used for the discovery of new concepts in streams of data.

Abstract

Abstract
Wireless networks play an increasingly important role in application areas such as factory-floor automation, process control, and automotive electronics. In this paper, we address the problem of sharing a wireless channel among a set of sporadic message streams where a message stream issues transmission requests with real-time deadlines. For this problem, we propose a collision-free wireless medium access control (MAC) protocol, which implements static-priority scheduling and supports a large number of priority levels. The MAC protocol allows multiple masters and is fully distributed; it is an adaptation to a wireless channel of the dominance protocol used in the CAN bus, a proven communication technology for various industrial applications. However, unlike that protocol, our protocol does not require a node having the ability to receive an incoming bit from the channel while transmitting to the channel. The evaluation of the protocol with real embedded computing platforms is presented to show that the proposed protocol is in fact collision-free and prioritized. We measure the response times of our implementation and find that the response-time analysis developed for the protocol indeed offers an upper bound on the response times.

Abstract

Abstract
A two-level data transformation consists of a type-level transformation of a data format coupled with value level transformations of data instances corresponding to that format. We have implemented a system for performing two-level transformations on XML schemas and their corresponding documents, and on SQL schemas and the databases that they describe. The core of the system consists of a combinator library for composing type-changing rewrite rules that preserve structural information and referential constraints. We discuss the implementation of the system's core library, and of its SQL and XML front-ends in the functional language Haskell. We show how the system can be used to tackle various two-level transformation scenarios, such as XML schema evolution coupled with document migration, and hierarchical-relational data mappings that convert between XML documents and SQL databases.

Abstract