Search results for "Computer architecture"
showing 10 items of 191 documents
Positionless aspect based sentiment analysis using attention mechanism.
2021
Abstract Aspect-based sentiment analysis (ABSA) aims at identifying fine-grained polarity of opinion associated with a given aspect word. Several existing articles demonstrated promising ABSA accuracy using positional embedding to show the relationship between an aspect word and its context. In most cases, the positional embedding depends on the distance between the aspect word and the remaining words in the context, known as the position index sequence. However, these techniques usually employ both complex preprocessing approaches with additional trainable positional embedding and complex architectures to obtain the state-of-the-art performance. In this paper, we simplify preprocessing by …
EV-planning: Electric vehicle itinerary planning
2014
International audience; In the latest few years, lot of efforts have been done to pave the way to sustainable mobility, in order to solve pollution problems and fuel shortage. The use of electric vehicles (EV) is considered as one of the best ecologic and economic solution. However, autonomy barriers and limitations slow the progress and the deployment of this technology. In this paper, we propose an advanced electric vehicles' fleet management architecture. This architecture considers the most important factors that can affect the traveling mode of electric vehicles, in order to offer different services to fleet management companies for an efficient monitoring and management of their fleet…
Forbidden Factors and Fragment Assembly
2002
In this paper we approach the fragment assembly problem by using the notion of minimal forbidden factors introduced in previous paper. Denoting by M(w) the set of minimal forbidden factors of a word w, we first focus on the evaluation of the size of elements in M(w) and on designing of an algorithm to recover the word w from M(w). Actually we prove that for a word w randomly generated by a memoryless source with identical symbol probabilities, the maximal length m(w) of words in M(w) is logarithmic and that the reconstruction algorithm runs in linear time. These results have an interesting application to the fragment assembly problem, i.e. reconstruct a word w from a given set I of substrin…
Words and Patterns
2002
In this paper some new ideas, problems and results on patterns are proposed. In particular, motivated by questions concerning avoidability, we first study the set of binary patterns that can occur in one infinite binary word, comparing it with the set of factors of the word. This suggests a classification of infinite words in terms of the "difference" between the set of its patterns and the set of its factors. The fact that each factor in an infinite word can give rise to several distinct patterns leads to study the set of patterns of a single finite word. This set, endowed with a natural order relation, defines a poset: we investigate the relationships between the structure of such a poset…
M-VIF: A machine-vision based on information fusion
2002
The authors describe a new architecture for machine vision, which is based on information fusion approach. Its general design has been developed by using a formal computation model that integrates three main ingredients of the visual computation: the data, the models, and the algorithms. The hardware design and the software environment of M-VIF are also given. The simulation of M-VIF is under development on the HERMIA-machine.
An A* Based Semantic Tokenizer for Increasing the Performance of Semantic Applications
2013
Semantic Applications (SAs) makes use of ontolo- gies and their performance can depend on the syntactic labels of the modeled entities; even if several approaches have been devised to formalize ontologies, no formal approaches have been devised for naming their constituents, which look as long word concatenations without any particular separation. We present a novel semantic tokenizer that finds the sub-words through an application of the A* based search algorithm; the A* functions rely on a set of linguistic criteria and on the meta-cognitive perspective of the activity of reading.
A Structural Approach to Infer Recurrent Relations in Data
2014
Extracting knowledge from a great amount of collected data has been a key problem in Artificial Intelligence during the last decades. In this context, the word "knowledge" refers to the non trivial new relations not easily deducible from the observation of the data. Several approaches have been used to accomplish this task, ranging from statistical to structural methods, often heavily dependent on the particular problem of interest. In this work we propose a system for knowledge extraction that exploits the power of an ontology approach. Ontology is used to describe, organise and discover new knowledge. To show the effectiveness of our system in extracting and generalising the knowledge emb…
Markov Model for Tweets Geographic Distribution Characterization
2015
Abstract In this paper we will continue our researches regarding e-Business and e-Government modeling on Social Media presented in (Stoica, Pitic, & Mihaescu, 2013). Among message and user parameters we add a new parameter used to describe the geographical dispersion of Twitter messages. This new parameter will characterize the way one set of messages will spread in Social Graph from the physical word point of view. The first model, presented as “A Novel Model for E-Business and E-Government Processes on Social”, will be extended with the geographical parameter PG. We will define and we will describe the Markov Model used to organize the messages gathered from social media. The main idea of…
Automated Integration and Communication Synthesis of Reconfigurable MPSoC Platform
2007
The communication synthesis is the main problematic in the multiprocessor system-on-chip (MPSoC). To resolve this problem, several methodologies can be used. These methodologies require automated methods to specify, generate and optimize the hardware, software, and the architectural interfaces between them. In this paper, we present a methodology flow for hardware-software communication synthesis for multiprocessor system-on-chip platform which are dedicated to streaming applications. Our methodology consists of high level architecture communication synthesis from functional description of the MPSoC design. The solution that we propose consists in synthesizing a custom bus architecture for …
Space information is important for reading
2009
AbstractReading a text without spaces in an alphabetic language causes disruption at the levels of word identification and eye movement control. In the present experiment, we examined how word discriminability affects the pattern of eye movements when reading unspaced text in an alphabetic language. More specifically, we designed an experiment in which participants read three types of sentences: normally written sentences, regular unspaced sentences, and alternatingbold unspaced sentences. Although there was a reading cost in the unspaced sentences relative to the normally written sentences, this cost was much smaller in alternatingbold unspaced sentences than in regular unspaced sentences.