Rare-aware attention network for image–text matching     

《Information processing & management》2023,60(3):103280

Image and text matching bridges visual and textual modality differences and plays a considerable role in cross-modal retrieval. Much progress has been achieved through semantic representation and alignment. However, the distribution of multimedia data is severely unbalanced and contains many low-frequency occurrences, which are often ignored and cause performance degradation, i.e., the long-tail effect. In this work, we propose a novel rare-aware attention network (RAAN), which explores and exploits textual rare content for tackling the long-tail effect of image and text matching. Specifically, we first design a rare-aware mining module, which contains global prior information construction and rare fragment detector for modeling the characteristic of rare content. Then, the rare attention matching utilizes prior information as attention to guide the representation enhancement of rare content and introduces the rareness representation to strengthen the similarity calculation. Finally, we design prior information loss to optimize the model together with the triplet loss. We perform quantitative and qualitative experiments on two large-scale databases and achieve leading performance. In particular, we conduct 0-shot test for rare content and improve rSum by 21.0 and 41.5 on Flickr30K (155,000 image and text pairs) and MSCOCO (616,435 image and text pairs), demonstrating the effectiveness of the proposed method for the long-tail effect.  相似文献   

Neural embedding-based indices for semantic search     

Fatemeh Lashkari  Ebrahim Bagheri  Ali A. Ghorbani 《Information processing & management》2019,56(3):733-755

Traditional information retrieval techniques that primarily rely on keyword-based linking of the query and document spaces face challenges such as the vocabulary mismatch problem where relevant documents to a given query might not be retrieved simply due to the use of different terminology for describing the same concepts. As such, semantic search techniques aim to address such limitations of keyword-based retrieval models by incorporating semantic information from standard knowledge bases such as Freebase and DBpedia. The literature has already shown that while the sole consideration of semantic information might not lead to improved retrieval performance over keyword-based search, their consideration enables the retrieval of a set of relevant documents that cannot be retrieved by keyword-based methods. As such, building indices that store and provide access to semantic information during the retrieval process is important. While the process for building and querying keyword-based indices is quite well understood, the incorporation of semantic information within search indices is still an open challenge. Existing work have proposed to build one unified index encompassing both textual and semantic information or to build separate yet integrated indices for each information type but they face limitations such as increased query process time. In this paper, we propose to use neural embeddings-based representations of term, semantic entity, semantic type and documents within the same embedding space to facilitate the development of a unified search index that would consist of these four information types. We perform experiments on standard and widely used document collections including Clueweb09-B and Robust04 to evaluate our proposed indexing strategy from both effectiveness and efficiency perspectives. Based on our experiments, we find that when neural embeddings are used to build inverted indices; hence relaxing the requirement to explicitly observe the posting list key in the indexed document: (a) retrieval efficiency will increase compared to a standard inverted index, hence reduces the index size and query processing time, and (b) while retrieval efficiency, which is the main objective of an efficient indexing mechanism improves using our proposed method, retrieval effectiveness also retains competitive performance compared to the baseline in terms of retrieving a reasonable number of relevant documents from the indexed corpus.  相似文献   

Self-derivation through memory integration: A model for accumulation of semantic knowledge     

《Learning and Instruction》2020

Semantic knowledge accumulates through explicit means and productive processes (e.g., analogy). These means work in concert when information explicitly acquired in separate episodes is integrated, and the integrated representation is used to self-derive new knowledge. We tested whether (a) self-derivation through memory integration extends beyond general information to science content, (b) self-derived information is retained, and (c) details of explicit learning episodes are retained. Testing was in second-grade classrooms (children 7–9 years). Children self-derived new knowledge; performance did not differ for general knowledge (Experiment 1) and science curriculum facts (Experiment 2). In Experiment 1, children retained self-derived knowledge over one week. In Experiment 2, children remembered details of the learning episodes that gave rise to self-derived knowledge; performance suggests that memory integration is dependent on explicit prompts. The findings support nomination of self-derivation through memory integration as a model for accumulation of semantic knowledge and inform the processes involved.  相似文献   

Unmasking text plagiarism using syntactic-semantic based natural language processing techniques: Comparisons,analysis and challenges     

Vani K.  Deepa Gupta 《Information processing & management》2018,54(3):408-432

The proposed work aims to explore and compare the potency of syntactic-semantic based linguistic structures in plagiarism detection using natural language processing techniques. The current work explores linguistic features, viz., part of speech tags, chunks and semantic roles in detecting plagiarized fragments and utilizes a combined syntactic-semantic similarity metric, which extracts the semantic concepts from WordNet lexical database. The linguistic information is utilized for effective pre-processing and for availing semantically relevant comparisons. Another major contribution is the analysis of the proposed approach on plagiarism cases of various complexity levels. The impact of plagiarism types and complexity levels, upon the features extracted is analyzed and discussed. Further, unlike the existing systems, which were evaluated on some limited data sets, the proposed approach is evaluated on a larger scale using the plagiarism corpus provided by PAN¹ competition from 2009 to 2014. The approach presented considerable improvement in comparison with the top-ranked systems of the respective years. The evaluation and analysis with various cases of plagiarism also reflected the supremacy of deeper linguistic features for identifying manually plagiarized data.  相似文献   

国际科研众筹项目的特征描述、主题分析及属性对比:基于Experiment平台的发现     

刘周颖  赵宇翔  巴志超  吴聪  朱庆华 《图书情报知识》2020,(3):71-82

[目的/意义]为了深入了解科研众筹平台中的项目情况,本文对科研众筹项目的基本特征、研究主题和不同类型科研众筹项目的属性对比进行研究。[研究设计/方法]首先对科研众筹的起源、价值和运作模式进行了梳理;随后爬取了科研众筹平台Experiment的项目信息,并对该平台中科研项目的发起者、背书者、项目记录、所属学科、资助情况等进行特征描述;基于LDA模型对科研众筹项目主题进一步细化,并对不同资金筹集和不同主题分类情况下的科研众筹项目的属性差异进行比较分析。[结论/发现]科研众筹主题主要集中在生物学和生态学两个领域,不同资金筹集达成情况的科研众筹项目在项目背书者数量、信息记录次数和讨论次数上具有显著差异;不同主题的科研众筹项目在其讨论次数上具有显著差异。同时,是否提供视频和是否参与资助挑战在不同资金筹集达成情况和不同主题的科研众筹项目上均具有显著差异。[创新/价值]深入分析了科研众筹平台中项目主题及属性对比情况,对科研众筹平台优化信息审核机制、社交机制和信息展示机制具有一定的参考价值。  相似文献   

主题内容频数相关分析与历代书法理论演替关系研究     

张捷  饶薇  王小伦 《情报科学》2006,24(12):1822-1827

针对古代文献引用方式的特殊性和无法运用常规引文分析及关键词分析方法进行研究的现状，本文以主题内容为检索单元以朝代为时间单元进行频数统计研究，将贯穿中国历代传统书法理论（书论）中的大量自然意象描述主题内容分为21类，并以此为指标进行分朝代统计，进而采用线性回归方法计算不同朝代书论之间的相关性，并以此作为各朝书论之间的传承关系。论文同时利用文史考证材料对所揭示的传承现象进行了分析、验证和阐释。  相似文献   

网络环境下的知识挖掘   总被引：9，自引：0，他引：9  

侯雅柟 《情报科学》2003,21(8):887-890

当前网络信息大爆炸，要从大量信息中获得所需知识就需要运用知识挖掘方法。本文首先对数据、信息、知识三个概念进行了区分，阐述了网络知识挖掘的概念及类型，并从数据仓库、语义网络和XML等底层信息加工组织方法上对网络知识挖掘进行探讨。  相似文献   

面向语义Web的本体学习研究     

张建文  王强 《科技广场》2006,(8):18-20

语义Web的创建需要一套共同的标准概念体系,即本体(Ontology)。本体的构造手段仍然是以手工构造为主,效率和准确率都非常低,很容易导致知识获取的瓶颈。本文给出一个半自动化的需人工干预的本体学习体系结构,采用平衡的协作建模方式来构造语义Web中的本体;介绍了基于以上体系结构的本体学习的处理过程,并讨论了领域概念抽取,概念之间关系的抽取等关键技术。  相似文献   

生产模式描述工具及其应用研究     

唐中君  陈荣秋  纪雪洪 《预测》2006,25(1):8-16

基于图形语言的优点,对能描述生产模式的复杂性、集成性和系统性特点的工具进行了研究。在提出生产模式的基本构成要素的基础上,首先阐述了现有图形语言工具描述生产模式时的局限性,然后以大量生产为例提出了一种从整体描述生产模式的图形语言工具“生产模式树”,并进一步提出了生产模式树应该满足的标准、绘制方法厦生产模式树的优点。最后,运用生产模式树提出了一个体现大量定制中观层理论体系的运作模型。  相似文献   

我国科技期刊DOI使用现状分析——以大气科学类2014年版CJCR核心期刊为例          下载免费PDF全文

申乐琳 《中国科技期刊研究》2016,27(6):624-628

【目的】找出大气科学类期刊DOI注册和应用中存在的问题并及时解决,充分发挥DOI在科技期刊的作用,普及DOI检索在科研中的应用。【方法】调查分析2014版《中国科技期刊引证报告（核心版）》大气科学类核心期刊使用DOI的现状,包括DOI的注册及标注、文后参考文献著录,以及DOI的解析情况。【结果】大气科学类期刊DOI注册较成熟,但是DOI的应用和宣传远远不够,特别表现在文后参考文献的著录、首页的推荐引用格式,以及解析不准确等方面。【结论】提出科技期刊应做好DOI注册的后期工作,并加大DOI的宣传力度,提高DOI解析的正确性,在DOI的应用方面早日与国际接轨。  相似文献