Yixue He
Community@heyixue511-creator · Anhui Polytechnic University
Anhui Engineering University
Agent Skills by Yixue He
Showing 17 vetted skills indexed across 1 GitHub repositories.
heyixue511-creator文献计量分析
Cleans bibliographic datasets and detects keyword bursts for CiteSpace-ready WoS exports.
heyixue511-creatorS08-人工复核学习
Extracts filtering rules from manual review decisions to refine literature screening strategies.
heyixue511-creatorS16-CiteSpace剪枝与网络分析参考
Reference CiteSpace pruning algorithms and network metrics for bibliometric network analysis.
heyixue511-creatorS03-研究背景问询
Collects research topic, discipline, and purpose through structured questions to define literature screening criteria.
heyixue511-creatorS15-基础数据分析
Computes statistics on cleaned bibliographic records including authors, journals, years, and keywords.
heyixue511-creatorS10-结果导出与日志
Exports cleaned bibliographic datasets, exclusion lists, and cleaning logs after each noise-removal round.
heyixue511-creatorS01-来源类型检测
Detects the source platform, export format, and encoding of bibliometric data files.
heyixue511-creatorS11-批判性审查
Audits literature cleaning results across six dimensions to detect systematic bias and misclassification.
heyixue511-creatorS09-噪音模式库
Maintains a reusable library of noise patterns discovered during bibliographic data cleaning.
heyixue511-creatorS04-映射表构建
Builds a CSV mapping table assigning unique IDs and search-term hit data to each bibliographic record.
heyixue511-creatorS07-分批语义判定
Classifies literature records as relevant or noise using semantic analysis and multi-round batch processing.
heyixue511-creatorS14-数据格式化导出
Converts cleaned bibliographic records into original WoS plain text format for CiteSpace.
heyixue511-creatorS02-来源数据分析
Analyzes bibliographic datasets to profile field completeness, document types, languages, and year distributions.
heyixue511-creatorS06-超长文本读取
Reads oversized bibliographic data files in record-aligned chunks to guarantee complete coverage.
heyixue511-creatorS13-关键词突现分析
Detect keyword bursts and research trends from cleaned Web of Science export files.
heyixue511-creatorS12-重复记录检测
Detects and resolves duplicate records in bibliometric datasets using DOI, title, and version matching.
heyixue511-creatorS05-检索策略解析
Analyzes bibliometric search queries to predict noise risks from wildcards and cross-domain homonyms.
Frequently Asked Questions About Yixue He
FAQPage SchemaWhat tasks can I accomplish with Yixue He's bibliometric skills?▼
You can detect literature data source formats, profile datasets, build mapping tables, semantically judge relevance per record, maintain a noise pattern library, deduplicate records, audit cleaning quality, and export CiteSpace-ready WoS files with keyword burst and network pruning analysis.
Who should use this bibliometric cleaning pipeline?▼
Academic researchers, graduate students, and bibliometricians preparing Web of Science literature datasets for CiteSpace visualization, systematic reviews, or research trend analysis who need rigorous, auditable semantic cleaning rather than simple keyword filtering.
How does the cleaning workflow run in practice?▼
The pipeline runs sequentially: S01 detects source type, S02 profiles data, S03 captures research context, S04 builds the mapping table, S05 parses search strategy risks, S07 performs batched semantic judgment, S10 exports results, and S14 formats final WoS output.
What are the prerequisites for using these skills?▼
You need a literature dataset exported from a platform such as Web of Science, a defined research topic for semantic anchoring, and optionally a search strategy string. Files over 2MB or 1000 records are handled via the S06 chunked reading skill.
How does the pipeline improve over repeated cleaning rounds?▼
The S08 manual review learning skill extracts knowledge from your correction decisions, infers optimization logic, and proposes rule revisions, while S09 maintains a coded noise pattern library with features, judgment methods, and boundary conditions reused across rounds.