the proceedings contain 42 papers. the topics discussed include: offensive text classification based on Ernie’s dual channel composite model;fake news detection algorithm based on incorporating multi-level features;E...
ISBN:
(纸本)9798400709241
the proceedings contain 42 papers. the topics discussed include: offensive text classification based on Ernie’s dual channel composite model;fake news detection algorithm based on incorporating multi-level features;EEGCN: event evolutionary graph comparison network for multi-modal fake news detection;explainable similar legal cases retrieval based on Siamese network;robust sentiment classification based on the backdoor adjustment;prior knowledge augmentation network for aspect-based sentiment analysis;a multi-dimension and multi-granularity feature fusion method for Chinese microblog sentiment classification;optimization study on weapon-target assignment problem based on intuitionistic fuzzy marine predator algorithm;a novel ranking method for textual adversarial attack;reinforcement learning in naturallanguageprocessing: a survey;and an application of co-plot analysis: a multidimensional scaling data visualization.
the proceedings contain 10 papers. the topics discussed include: predicting themes within complex unstructured texts: a case study on safeguarding reports;supporting ontology maintenance with contextual word embedding...
the proceedings contain 10 papers. the topics discussed include: predicting themes within complex unstructured texts: a case study on safeguarding reports;supporting ontology maintenance with contextual word embeddings and maximum mean discrepancy;mitigating the impact of out of vocabulary words in a neural-machine-translation-based question answering system;thesaurus enhanced extraction of Hohfeld’s Relations from Spanish labor law;extractive summarization for explainable sentiment analysis using transformers;deep learning enhanced with graph knowledge for sentiment analysis;explaining sentiment from lexicon;and semantic data transformation.
Reinforcement learning (RL) is a powerful technique for learning from data and feedback, but its effective application to naturallanguageprocessing (NLP) tasks remains an open question. Consequently, this paper firs...
详细信息
With rapid urbanization and increasing population density, effective waste sorting has become crucial for environmental protection. the diverse nature of waste-varying in type, shape, and background-necessitates preci...
详细信息
this paper evaluates the application of Artificial Intelligence algorithms for the legal document classification. Algorithms are divided into linear classifiers SVM and Logistic Regression, ULMFiT language Model and H...
详细信息
ISBN:
(纸本)9798350349122;9798350349115
this paper evaluates the application of Artificial Intelligence algorithms for the legal document classification. Algorithms are divided into linear classifiers SVM and Logistic Regression, ULMFiT language Model and HAN. the studied dataset is called VICTOR, composed of documents from the Brazilian Supreme Federal Court (STF). the article concludes that all machinelearning algorithms tested can be applied to classify legal documents from the employed dataset. Additionally, despite being less complex, the TF-IDF together linear classifiers outperform the experimented language Model and HAN in terms of F1-score.
Sound Symbolism is a well studied psychological phenomena of the relation between sound and meaning in naturallanguage. though the phenomena has been studied by psychologists and linguists, the phenomena has not been...
详细信息
ISBN:
(纸本)9798350349122;9798350349115
Sound Symbolism is a well studied psychological phenomena of the relation between sound and meaning in naturallanguage. though the phenomena has been studied by psychologists and linguists, the phenomena has not been put to use naturallanguageprocessing or modeled by machinelearning. In this work we select words for round and sharp objects from various naturallanguages. We attempted to see if a machinelearning algorithm could perform better than Chance in distinguishing words for round and sharp objects in naturallanguages. We performed a psychophysics experiment to see if human subjects will associate words for sharp objects with a round object and round object with sharp figure. We show that human subjects are more likely than chance to associate words for sharp objects with sharp figure and vice versa. We propose that the algorithms can be improved by using training sets consisting of words whose sound symbolic properties are labelled by psychophysics experiments.
the increasing volume of social media content has made hashtag recommendation a critical task for enhancing content discoverability and user engagement. this paper presents MRLKG (Multimodal Representation learning wi...
详细信息
the integration of the contrastive learning paradigm into deep clustering has led to enhanced performance in image clustering. However, in existing researches, the samples in the class of the target may be still treat...
详细信息
An innovative model on SVM is proposed, utilizing an enhanced version of the Dung beetle optimizer (DBO) algorithm in conjunction with Support Vector machine (SVM) optimization. the enhanced DBO algorithm incorporates...
详细信息
ISBN:
(纸本)9798350349122;9798350349115
An innovative model on SVM is proposed, utilizing an enhanced version of the Dung beetle optimizer (DBO) algorithm in conjunction with Support Vector machine (SVM) optimization. the enhanced DBO algorithm incorporates a Logistic chaotic mapping technique to improve population initialization, enhancing diversity and randomness. To further enhance its performance, strategies like Cauchy mutation and adaptive dynamic weighting are adopted to prevent premature convergence and local optima. this improved algorithm is then applied to SVM parameter optimization, resulting in improved classification performance. To validate the effectiveness of the proposed method, a model called HDBO-SVM is constructed and tested on the UCI dataset. the experimental results substantiate the outstanding performance of the HDBO-SVM model, showcasing a significant improvement compared to conventional models.
this paper presents a novel task-short interest trend prediction. We provide its formal definition and create a dataset for this task. In addition to a thorough dataset analysis, to explore the possibility of automati...
详细信息
ISBN:
(纸本)9798350349122;9798350349115
this paper presents a novel task-short interest trend prediction. We provide its formal definition and create a dataset for this task. In addition to a thorough dataset analysis, to explore the possibility of automating this task, we present the results of the majority baseline, machinelearning models, and neural network models. Our results indicate that this task can be challenging, and even recently pre-trained large neural networks still struggle to tackle this task.
暂无评论