This paper introduces the Bank Question (BQ) corpus, a Chinese corpus for sentence semantic equivalence identification (SSEI). The BQ contains 120,000 question pairs from 1-year online bank custom service logs. To efficiently process and annotate questions such large scale of logs, this proposes clustering based annotation method to achieve with same intent. First, deduplicated answer are clustered into stacks by Word Mover’s Distance (WMD) Affinity Propagation (AP) algorithm. Then,...

Rice (Oryza sativa L.), a major staple food for billions of people, was assessed its phytotoxicity copper oxide nanoparticle (CuO NPs, size < 50 nm). Under hydroponic condition, seven days exposure to 62.5, 125, and 250 mg/L CuO NPs significantly suppressed the growth rate rice seedlings compared both control treatment supernatant from NP suspensions. In addition, physiological indexes associated with antioxidants, including membrane damage antioxidant enzyme activity, were also detected....

Rapid urbanization and associated transportation play a crucial role in the distribution of heavy metals road dust, leading to serious environmental health concerns. This study explored concentration spatial variability metals, including cadmium (Cd), copper (Cu), lead (Pb), zinc (Zn), dust across six urban zones Guangdong Province (parks, educational, residential, commercial, traffic, industrial areas). The also evaluated bioavailability these determine their carcinogenic non-carcinogenic...

University libraries have archaeologically augmented scientific research by collecting, organizing, maintaining, and availing materials for access. Researchers reckon that with the expertise acquired from conventional cataloging, classification, indexing coupled attained in development, along maintenance of institutional repositories, it is only rational take a dominant central role data management further their capacity as curators. Accordingly, are expected to assemble capabilities, manage...

We present a large-scale dataset for the task of rewriting an ill-formed natural language question to well-formed one. Our multi-domain (MQR) is constructed from human contributed Stack Exchange edit histories. The contains 427,719 pairs which come 303 domains. provide annotations subset as quality estimate. When moving questions, improves by average 45 points across three aspects. train sequence-to-sequence neural models on and obtain improvement 13.2% in BLEU-4 over baseline methods built...

The contradictions between the working times and learning time of adult in evening university correspondence education have become more serious. Anecdotal evidence suggests that traditional educations (There are no online courses) not able to ease contradiction. It has been found course can solve conflict students education. Therefore, many academies which engaged brought courses model. However, how ensure implementary quality a vital problem, paper, Online Course Quality Maturity Model...

Abstract Automated essay scoring ( AES ) generally computes scores as a function of macrofeatures derived from set microfeatures extracted the text using natural language processing NLP ). In e‐rater ® automated engine, developed at Educational Testing Service ETS for essays, each four grammar, usage, mechanics , and style [ GUMS ]) is computed microfeatures. Statistical analyses reveal that some these might not explain much variance in human regardless writing tasks. Currently, same...

The aim of this study was to examine the effect non‐pharmacological staged interventions on fatigue and dyspnoea in patients with chronic obstructive pulmonary disease. A randomized controlled trial conducted 64 a tertiary hospital C hina from 2010 2011. Patients were randomly assigned control group ( n = 32), who received routine care, intervention additional interventions. M ultidimensional F atigue I nventory five‐grade edical R esearch ouncil scale used collect data at baseline after 6...

This paper describes our participation in SemEval 2018 Task 2: Multilingual Emoji Prediction, which participants are asked to predict a tweet’s most associated emoji from 20 emojis. Instead of regarding it as 20-class classification problem we regard text similarity problem. We propose vector based approach for this task. First the distributed representation (tweet vector) each tweet is generated, then between and emoji’s embedding evaluated. The similar chosen predicted label. Experimental...

In view of problems errors in the nonlinear measurement existing electronic scales and temperature change, a new scale measuring system is designed. When bridge strain gauge deformation occurring, will change analog quantity. Then, quantity transformed into digital data are sent to MCU. After that further processed by software filtering. Finally, realizes functions weight objects, saving price information, peeling displaying through LCD. Compared with traditional system, this has higher...

Nowadays, adaptive e-learning systems are responsible for selecting learning materials or contents according to the learners' style, profile, interest, previous knowledge level, goal, pedagogical method etc provide highly personalized sessions. Complex system (CAS) lays stress on adaptation and positive feedbacks. And semantic Web service aims at giving information a well-defined meaning, enabling computers people work in cooperation better through definition of ontologies. This paper...

Adaptive e-learning system, deal with appropriate personalization and adaptation techniques in order to maximize the effectiveness of learning. Customer preference personality are going be more important e-learning. This paper first hopes explain grid agent model, whose main actions including registry, directory discovery. Through these actions, managerpsilas will find out suitable learning services. Secondly, implement adaptability method artificial psychology how realize adaptive...

A novel stereo video matching algorithm based on multiple searching windows in rank transform domain is proposed this paper. By doing a transform, videos are mapped to the space from gray-level space, which eliminate interference of environment variations and diverse cameras. The disparity estimated by multi-window with fast strategy, improves accuracy depth map speed as well. After that, compensation step done smooth area get smoother clearer region objects. experiment results show...

In the era of digital transformation, integration information technology into learning has posed challenges and opportunities. Navigating global knowledge economy requires individuals to acquire skills facilitating adaptation evolving economic societal landscapes. Deep learning, recognized as pivotal for addressing real-world fostering adaptive capabilities, garnered attention in educational research. This study investigates effect a deep teaching approach on academic achievement computer...

Just noticeable distortion (JND), representing the threshold of in an image that is minimally perceptible to human visual system (HVS), crucial for compression algorithms achieve a trade-off between transmission bit rate and quality. However, traditional JND prediction methods only rely on pixel-level or sub-band level features, lacking ability capture impact content JND. To bridge this gap, we propose Semantic-Guided (SG-JND) network leverage semantic information prediction. In particular,...

To meet the development and application needs of advanced high-strength steel, laser welding 22MnB5 hot-forming steel plates with thicknesses 2 mm, 3 4 mm was studied in this paper. Mechanical testing revealed that as plate thickness increased, tensile strength welded joints decreased from 1489 MPa to 1357 1275 MPa, equating 96%, 91%, 88% corresponding base metal strength, respectively. The heat-affected zone exhibited lowest mechanical properties. Microstructural characterization showed...