Part II Knowledge Engineering Research and Applications (KERA) -- Pretrain-Transfer Strategy -- English Cloze Test Based on BERT -- An Automatic Method for Understanding Political Polarization through Social Media -- An Improved Convolutional Neural Network Based on Noise Layer -- Syntactic Enhanced Projection Network for Few-shot Chinese Event Extraction -- A Framework of Data Augmentation While Active Learning for Chinese Named Entity Recognition -- Traffic Route Planning in Partially Observable Enviroment Using Actions Group Representation -- Bayesian belief network Model using Sematic Concept for Expert Finding -- Towards Solving the Winograd Schema Challenge: Model-Free, Model-Based and a Spectrum in Between -- The Novel Efficient Transformer for NLP -- Pretrain-Transfer Strategy -- English Cloze Test Based on BERT -- An Automatic Method for Understanding Political Polarization through Social Media -- An Improved Convolutional Neural Network Based on Noise Layer -- Syntactic Enhanced Projection Network for Few-shot Chinese Event Extraction -- A Framework of Data Augmentation While Active Learning for Chinese Named Entity Recognition -- Traffic Route Planning in Partially Observable Enviroment Using Actions Group Representation -- Bayesian belief network Model using Sematic Concept for Expert Finding -- Towards Solving the Winograd Schema Challenge: Model-Free, Model-Based and a Spectrum in Between -- The Novel Efficient Transformer for NLP -- Joint Entity and Relation Extraction for Long Text -- Evaluating dataset creation heuristics for concept detection in web pages using BERT.-Panoptic-DLA: Doucument Layout Analysis of Historical Newspapers Based on Proposal-Free Panoptic Segmentation Model -- Improving Answer Type Classification Quality Through Combined Question Answering Datasets -- FOBA: Flight Operation Behavior Analysis Based On Hierarchical Encoding -- An Event Detection Method Combining Temporal Dimension and Position Dimension -- Local Feature Normalization -- Combining Knowledge with Attention Neural Networks for Short Text Classification -- A Dialogue Contextual Flow Model for Utterance Intent Recognition in Multi-turn Online Conversation -- An Empirical Study on Effect of Semantic Measures in Cross-domain Recommender System in User Cold-start Scenario -- Community Enhanced Course Concept Recommendation in MOOCs with Multiple Entities -- AABC:ALBERT-BiLSTM-CRF combining with Adapters -- Q-learning with Fisher Score for Feature Selection of Large-scale Data Sets -- The Modularity of Inconsistent Knowledge Bases with Application to Measuring Inconsistency -- Collaborative Embedding for Knowledge Tracing -- Construction and Analysis of Cross-Regional Emergency Collaboration Network Model -- Residual gated recurrent unit-based stacked network for stock trend prediction from limit order book -- A Social Attribute Inferred Model Based on Spatio-Temporal Data -- Chinese Judicial Abstracting Based on Short Sentence Extraction and GPT-2 -- Extracting Anomalous Pre-Earthquake Signatures from Swarm Satellite Data using EOF and PC analysis -- Hierarchical Multi-label Text Classification: Self-adaption Semantic Awareness Network Integrating Text Topic and Label Level Information -- Sememes-based Framework for Knowledge Graph Embedding with Comprehensive-Information -- Domain-Specific Sentence Encoder for Intention Recognition in Large-Scale Shopping Platforms -- Chinese Event Detection Based on Event Ontology and Siamese Network -- Alzheimer's Disease Prediction Using EfficientNet and Fastai -- Interval Occlusion Calculus with Size Information -- Adaptive Entity Alignment for Cross-lingual Knowledge Graph.-Similarity-Based Heterogeneous Graph Attention Network for Knowledge-Enhanced Recommendation -- Accurate and robust RGB-D visual odometry based on point and line features -- Sentence Extraction-Based Machine Reading Comprehension for Vietnamese -- Analyzing and Recommending Development Order based on Design ClassDiagram -- HCapsNet: A Text Classification Model Based On Hierarchical Capsule Network -- Sentence Matching With Deep Self-Attention and Co-Attention Features -- Not Only the Contextual Semantic Information ï¼A Deep Fusion Sentimental Analysis Model towards Extremely Short Comments -- A Label Noise Robust Cross-Modal Hashing Approach -- Acoustic Modeling for Indoor Spaces Using Ray-Tracing Method -- Learning path recommendation for MOOC platforms based on a knowledge graph -- Spy The Lie: Fraudulent Jobs Detection in Recruitment Domain using Knowledge Graphs -- Performance Evaluation of Multi-class Sentiment Classification using Deep Neural Network Models Optimised for Binary Classification.-Discovering Stable Ride-sharing Groups for Commuting Private Car using Spatio-temporal Semantic Similarity -- SA2SL: From Aspect-Based Sentiment Analysis to Social Listening System for Business Intelligence. .

This three-volume set constitutes the refereed proceedings of the 14th International Conference on Knowledge Science, Engineering and Management, KSEM 2021, held in Tokyo, Japan, in August 2021. The 164 revised full papers were carefully reviewed and selected from 492 submissions. The contributions are organized in the following topical sections: knowledge science with learning and AI; knowledge engineering research and applications; knowledge management with optimization and security.