Classification methods for streaming data are not new, but very few current frameworks address all three of the most common problems with these tasks: concept drift, noise, and exorbitant costs associated labeling unlabeled instances in streams. Motivated by this gap field, we developed an active learning framework based on a dual-query strategy Ebbinghaus's law human memory cognition. Called CogDQS, query samples only representative manual annotation local density uncertainty, thus...

Few-shot relation extraction constitutes a critical task in natural language processing. Its aim is to train model using limited number of labeled samples when data are scarce, thereby enabling the rapidly learn and accurately identify relationships between entities within textual data. Prototypical networks extensively utilized for simplicity efficiency few-shot scenarios. Nevertheless, prototypical derive their prototypes by averaging feature instances given category. In cases where...

HDX mass spectrometry is a powerful platform to probe protein structure dynamics during ligand binding, folding, enzyme catalysis, and such. analysis derives the based on increase of which backbone protons exchanged with solvent deuterium. Coupled digestion MS/MS analysis, can be used study regional m/z value or percentage deuterium incorporation for digested peptides in experiments. Various software packages have been developed analyze data. Despite progresses, proper explicit statistical...

This paper considers a practical scenario—data stream classification, and uses online active learning to assist the classifier. Most methods are proposed based on label uncertainty, but representativeness faced with difficulties in scenarios. The sampling criteria usually fixed, which makes it difficult for them work effectively dynamic sample spaces, especially data streams concept drift. proposes novel framework representativeness. local nearest-neighbor information measure of unlabeled...

Abstract Background Multidimensional protein identification technology (MudPIT)-based shot-gun proteomics has been proven to be an effective platform for functional proteomics. In particular, the various sample preparation methods and bioinformatics tools can integrated improve applications like target organelle We have recently a rapid method classification system comparative analysis of plant responses two hormones, zeatin brassinosteroid (BR). These hormones belong distinct classes growth...

Enzyme dynamics has recently been shown to be crucial for structure-function relationship. Among various structure analysis platforms, HDX (hydrogen deuterium exchange) mass spectrometry stands out as an efficient and high-throughput way analyze protein upon ligand binding. Despite the potential, limited research employed spec platform probe regional of enzymes. In particular, technique never used analyzing cell wall degrading We hereby xylanase a model explore potential studying revealed...

In grid computing, group communication is an important strategy to realize large-scale information resource sharing. However, it very difficult ensure the security of in environment. this paper, based on basic theories threshold signature and characteristics grid, we present four algorithms, including keys generating, individual generating verifying, encrypting, decrypting which constitute authenticated encryption mechanism for grid. Finally, validate correctness proposed paper analyze its...

Abstract Entity synonyms play a significant role in entity-based tasks. Previous approaches use linguistic syntax, distributional, and semantic features to expand entity synonym sets from text corpora. Due the flexibility complexity of Chinese language expression, aforementioned are still difficult robustly text, because these fail track holistic semantics among entities suffer error propagation. This paper introduces an approach for expanding based on bilateral context filtering strategy....

The accuracy of autonomous navigation and obstacle avoidance unmanned aerial vehicles (UAVs) in complex environments has become one challenging task. In this paper, an the UAV (ANOAU) algorithm based on deep reinforcement learning (DRL) been proposed to achieve accurate path planning environments. our work, we use actor–critic‐based DRL framework control from sensor input output UAV’s action design a set reward functions that can be adapted for environment. Meanwhile, alleviate...

Cooperative processing is a representative application in pervasive grid and group communication the key technique this application. It great challenge to ensure security of grid. In paper, we first describe service infrastructure middleware for Then, present secure mechanism analyse correctness mechanism. Finally, verify validity by experiments. Results show that our proposed efficient

This work develops a construction pattern for the wireless sensor network. While existing works have developed various methods water quality detection tasks, they primarily focused on establishment of network in single area to collect and transmit data general overlooked combination multiple areas, which has geographical limitations. As such, these can be suboptimal monitoring quality. In this paper, we propose new framework named WQD, short detection. Firstly, perform detailed feature...

<title>Abstract</title> Class imbalance inevitably occurs in dynamic data stream scenarios and can pose tremendous challenges for mining. To address these challenges, an adaptive resampling weighted ensemble method (ARWE) is proposed this paper. First, the subdivision Poisson (DSPR) module ARWE developed to class problem thedata stream. DSPR combines local information from minority samples with rate design a sample-weighting scheme that enhance visibility of samples, particularly those at...

Traversing Network Address Translation (NAT) for Peer-to-Peer (P2P) communication has become a hot topic recently. Compared to UDP, establishing TCP connections hosts behind different NATs is more complex. Thus, many TCP-based applications do not address traversal through NATs. Some solutions suggest using delegates relay all communications, or tunneling over UDP. However, they require big reform network architecture, non-standard TCP/IP stack. In this paper, we present novel idea called...

This article mainly research on energy regeneration for the electric vehicle, analyze of vehicles in process braking, how we can recovery more efficient generated during braking. To calculate efficiency by a mathematical model based loss vehicle motors, obtain corresponding parameters when is best.

Data stream classification becomes a promising prediction work with relevance to many practical environments. However, under the environment of concept drift and noise, research data faces lots challenges. Hence, new incremental ensemble model is presented for classifying nonstationary streams noise. Our approach integrates three strategies: learning monitor adapt drift; improve stability; microclustering procedure that distinguishes from noise predicts labels incoming instances via majority...

Entity synonym sets play a significant role in many entity-based tasks. Robustly discovering entity requires analyzing the linguistic and semantic features of entities target language. Although there are approaches to expand from English text, it is still difficult robustly Chinese text because flexibility complexity language expression. In this paper, we propose an approach for expanding via bilateral context filtering strategies. Specifically, mainly includes three parts. First, gains...