A5007434394- Natural Language Processing Techniques
- Software System Performance and Reliability
- Topic Modeling
- Intelligent Tutoring Systems and Adaptive Learning
- Tuberculosis Research and Epidemiology
- Cloud Computing and Resource Management
- Urban Transport and Accessibility
- Simulation and Modeling Applications
- Service-Oriented Architecture and Web Services
- Evaluation Methods in Various Fields
- Mycobacterium research and diagnosis
- Geological Modeling and Analysis
- Pneumocystis jirovecii pneumonia detection and treatment
- Endometrial and Cervical Cancer Treatments
- Open Education and E-Learning
- Face and Expression Recognition
- Grey System Theory Applications
- Urban Green Space and Health
- Image Processing Techniques and Applications
- Urban and Freight Transport Logistics
- Parallel Computing and Optimization Techniques
- Food Supply Chain Traceability
- Face recognition and analysis
- Safety and Risk Management
- Urban Stormwater Management Solutions
University of Science and Technology of China
2024
Shandong Jiaotong University
2024
Shandong Transportation Research Institute
2024
Hunan University
2019-2023
Hunan Normal University
2021-2022
Sun Yat-sen University
2021-2022
Zhejiang Provincial Public Security Department
2022
Northeastern University
2021
Beijing Normal University
2021
Bridge University
2021
As the first and foremost step of typical automatic log analysis, parsing has attracted a lot interest. Most existing studies treat messages as pure strings rely on string matching or distance. In NLP, word2vec shown very efficient effective in representing words with low dimensional vectors. Inspired by this, this paper we propose novel method, called LPV (Log Parser based Vectorization), for both offline online parsing. The central idea our method is to convert into vectors, measure...
Detecting anomalies in logs is crucial for service and system management, since are widely used to record the runtime status, often only data available postmortem analysis. Since usually rare real-world services systems, a common feasible practice mine or learn normal patterns from logs, deem those violating as anomalies. As log sequences kind of time series data, RNN (Recurrent Neural Network) its variants have been extensively employed capture patterns. Nevertheless, sequential nature...
Object-oriented change detection (OOCD) plays an important role in remote sensing detection. Generally, most of current OOCD methods adopt the highest predicted probability to determine whether objects have changes. However, it ignores fact that only parts object changes, which will generate uncertain classification information. To reduce uncertainty, improved rough-fuzzy possibilistic <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Exploring the heterogeneous characteristics of urban expansion process is essential for understanding dynamics spatial structure. Many studies focused on depicting spatio-temporal based patches. However, measuring from agglomeration areas comprising expanded construction land patches have not been adequately explored. This study presents a novel approach and two improved indices characterizing heterogeneity during expansion. Firstly, we proposed Gaussian mixture model considering multiple...
The reasonable layout of cold chain logistics base facilities in Jinan is an important path to promote the healthy development logistics. location selection must be closely related economic surrounding area and industrial planning city, cannot separated from environment. At same time, close consumer market, which can reduce transportation cost, expand radiation radius increase purchasing power consumers. This paper first explains City, then analyzes site based on Analytic Hierarchy Process...
Geometry Problem Solving (GPS), which is a classic and challenging math problem, has attracted much attention in recent years. It requires solver to comprehensively understand both text diagram, master essential geometry knowledge, appropriately apply it reasoning. However, existing works follow paradigm of neural machine translation only focus on enhancing the capability encoders, neglects characteristics human In this paper, inspired by dual-process theory, we propose Dual-Reasoning Solver...
Rifampin is the first-line antituberculosis drug, with Mycobacterium tuberculosis RNA polymerase as molecular target. Unfortunately, M. strains that are resistant to rifampin have been identified in clinical settings, which limits its therapeutic effects. In isolates, S531L and D516V (in Escherichia coli ) two common mutated codons gene rpoB , corresponding S456L D441V .
Logs are pervasive in modern computing systems, and valuable to service system management. Nevertheless, with the rapidly growing size complexity of log volume is exploding, which makes automatic analysis imperative. Generally, analysis, first fundamental step parsing, a lot effort has been devoted. However, most existing parsing methods, messages merely treated as plain text. In natural language processing (NLP) area, it common practice represent words sentences vectors, then similarity...
Math Word Problems (MWPs) are crucial for evaluating the capability of Large Language Models (LLMs), with current research primarily focusing on questions concise contexts. However, as real-world math problems often involve complex circumstances, LLMs' ability to solve long MWPs is vital their applications in these scenarios, yet remains under-explored. This study pioneers exploration Context Length Generalizability (CoLeG), LLMs MWPs. We introduce Extended Grade-School (E-GSM), a collection...
Educational resource understanding is vital to online learning platforms, which have demonstrated growing applications recently. However, researchers and developers always struggle with using existing general natural language toolkits or domain-specific models. The issue raises a need develop an effective easy-to-use one that benefits AI education-related research applications. To bridge this gap, we present unified, modularized, extensive library, EduNLP, focusing on educational...
Aligning Large Language Models (LLMs) with human intentions and values is crucial yet challenging. Current methods primarily rely on preferences, which are costly insufficient in capturing nuanced feedback expressed natural language. In this paper, we present Self-Refinement Tuning (SRT), a method that leverages model for alignment, thereby reducing reliance annotations. SRT uses base language (e.g., Tulu2) to generate initial responses, critiqued refined by more advanced GPT-4-Turbo). This...
Abstract Background: Equitable access to sports services has drawn attention from policymakers and planners in China as people’s health been placed at the centre of country’s policy-making machinery. However, existing approaches measure spatial accessibility facilities tend ignore heterogeneity potential users’ demands facility preferences, thereby causing a bias measurement accessibility. Methods: To accurately facilities, this paper proposes multi-preference Gaussian two-step floating...
Impervious surfaces are essential elements for the urban ecological environment. Machine-learning-based approaches have achieved successful breakthroughs in impervious surface extraction. These methods require large sets of labeled data to train a model. However, it is challenge acquire massive sample because complexity, time consumption, and high cost. To address this issue, we explore method generate training samples using point interest (POI) vehicle trajectory global positioning system...