Manually labeling documents for training a text classifier is expensive and time-consuming. Moreover, trained on labeled may suffer from overfitting adaptability problems. Dataless classification (DLTC) has been proposed as solution to these problems, since it does not require documents. Previous research in DLTC used explicit semantic analysis of Wikipedia content measure distance between documents, which turn classify test based nearest neighbours. The semantic-based method major drawback...

A reliable and accurate battery model is the basis of prediction voltage state power (SOP). Based on electrochemical a battery, an improved novel polarization (NPV) based current time developed in this study. The parameters NPV can be identified with only small-batch primary data, SOP at t-second (t > 0) under constant (I ≠ easily realized. simulation results show that when 3C used to charge directly from = 0 cut-off voltage, errors terminal are 1.4% 4.9%, respectively, average them whole...

Due to its low cost, environmental friendliness and high energy density, the lithium-sulfur battery (LSB) has been regarded as a promising alternative for next generation of rechargeable systems. However, practical application LSB is seriously hampered by short cycle life self-charge owing apparent shuttle effect soluble lithium polysulfides. Using MgSO4@MgO composite both template dopant, template-guided S-doped mesoporous graphene (SMG) prepared via fluidized-bed chemical vapor deposition...

Restricted Boltzmann machines (RBMs) and their variants are usually trained by contrastive divergence (CD) learning, but the training procedure is an unsupervised learning approach, without any guidances of background knowledge. To enhance expression ability traditional RBMs, in this paper, we propose pairwise constraints (PCs) RBM with Gaussian visible units (pcGRBM) model, which guided PCs process encoding conducted under these guidances. The encoded hidden layer features pcGRBM. Then,...

Abstract Isolated Sign Language Recognition (ISLR), which seeks to automatically align sign videos with corresponding glosses, has recently gained considerable attention from the artificial intelligence community. This technology potential bridge communication gap between hearing people and deaf However, development of ISLR is hindered by scarcity language datasets. Moreover, existing datasets are limited their provision a single perspective, makes hand gesture occlusion difficult handle. In...

This paper addresses the problem of anomaly detection for high-dimensional sensing data. The one-class support vector machine (OCSVM) is one most popular unsupervised methods detection. When data are high dimensional and large scale, however, efficiency OCSVM-based in suffers. Although dimensionality-reduction tools, such as deep belief networks, can be applied to compress alleviate problem, accuracy timely still hard improve due inherent features OCSVM. In this paper, we propose a new form...

Chitosan (CS) is a natural and low-cost adsorbent for capturing metal ions organic compounds. However, the high solubility of CS in acidic solution would make it difficult to recycle from liquid phase. In this study, CS/Fe3O4 was prepared via Fe3O4 nanoparticles immobilized onto surface, DCS/Fe3O4-Cu further fabricated after surface modification adsorption Cu ions. The meticulously tailored material displayed sub-micron size an agglomerated structure with numerous magnetic nanoparticles....

The Multilingual Chinese-English lexical sample task at SemEval-2007 provides a framework to evaluate Chinese word sense disambiguation and promote research. This paper reports on the preparation results of six participants.

Partnership for a new generation of vehicles (PNGV) model is conventional battery equivalent circuit (ECM). However, identifying the best parameters this challenge. In study, PNGV transformed into directly identifiable difference equation to identify its parameters. Subsequently, reference adaptive system (MRAS) used realize real-time identification The accuracy MRAS found be superior that recursive extended least square algorithm. For single hybrid pulse power characterization (HPPC),...

Accurate measurement of the open-circuit voltage (OCV) promotes state charge (SOC) accuracy. In this study, three transformation methods are employed to make OCV identifiable, and factors affecting accuracy identification investigated. Furthermore, a fast method is proposed. The results show that forward difference adaptive differential evolution algorithm more suitable for identification. affected by pulse characteristics, sampling frequency, C-rate, resting time between pulses....

With the advent and rapid development of image tampering technology, it has become harmful to many aspects our society. Thus, detection been increasingly important. Although current forgery methods have achieved some success, scale tampered areas in each are different, previous do not take this into account. In paper, we believe that inability network accommodate regions various sizes is main reason for low precision. To address mentioned problem, propose a neural architecture called...

Various cost-effective polycyclic aromatic hydrocarbons (PAHs) were used to fabricate hyper-cross-linked polymers (HCLPs) via an external cross-linker knitting method (ECLKM) followed by N-source impregnation modification. Multiple characterization techniques and thorough tests confirmed that the resultant thermally stable materials featured large specific surface areas (up 2870 m2 g–1), narrow pore distributions (<0.70 nm), high volumes 1.09 cm3/g). Effects of porosities N-doping...

With the rise of social media and internet, rapid dissemination information has increased likelihood reputation infringement. This study utilizes judicial big data AI to analyze intrinsic connections in infringement cases, aiding judges delivering consistent rulings. The challenge lies balancing freedom speech with right addressing ambiguity subjectivity cases. research constructs a structured case dataset from Chinese Judgments Online. It introduces Fractional Fuzzy Neural System (FFNS)...

This article addresses the problem of outlier detection for wireless sensor networks. As increasing amounts observational data are tending to be high-dimensional and large scale, it is becoming increasingly difficult existing techniques perform accurately efficiently. Although dimensionality reduction tools (such as deep belief network) have been utilized compress support detection, these methods may not achieve desired performance due special distribution compressed data. Furthermore,...

Cross-lingual document clustering is the task of automatically organizing a large collection multi-lingual documents into few clusters, depending on their content or topic. It well known that language barrier and translation ambiguity are two challenging issues for cross-lingual representation. To this end, we propose to represent through statistical word senses, which discovered from parallel corpus novel sense induction model method. In particular, former consists in sense-based vector...

This paper chose the Dual Polarization equivalent circuit model to simulate dynamic characteristics of lithium-ion battery. The relationship between state charge and open voltage battery was obtained based on an improved Hybrid Pulse Power Characteristic test. And then parameters in were identified effectively. established Simulink, verified constant current conditions respectively. results showed that can accurately reflect static battery, have high precision.