Day 1 (2024-04-24)
You always need to write something when publishing an article.
The following content is from my public account article (original Markdown):
Tip-Using AI-generated search queries for professional searching on academic websites like PubMed#
#written_by_10k
When conducting literature searches on platforms such as PubMed and PubScholar, it is crucial to standardize the search terms. For example, when searching for studies on the topic of "cancer," simply entering cancer in the search box is not sufficient for accurate retrieval. Although in some cases, the platform may automatically provide keyword expansion, there are two reasons why this may not be ideal:
-
Limitations of platform intelligence:
The level of platform intelligence is limited, which may result in unexpected search results during tokenization and other processes.
Search engines may introduce irrelevant or redundant search results when dealing with tokenization and other complex situations. -
Semantic understanding bias:
Platforms may fail to accurately understand the query meaning, resulting in a reduced number of articles retrieved.
Search engines may not fully understand user intent, leading to the omission of some relevant literature.
The above issues may cause discrepancies between the search results and expectations. In such cases, utilizing the advanced search function is more reliable and precise.
However, mastering and utilizing advanced search to construct complex search queries can be a relatively tedious task. Fortunately, AI technology can greatly assist in this aspect. Typically, any AI system integrated into such academic platforms is capable of handling search query construction:
Kimi: Below is a detailed explanation of the previously provided PubMed search queries:
-
("Curcumin"[Title/Abstract] OR "Turmeric"[Title/Abstract]) AND ("Review"[Publication Type] OR "Meta-Analysis"[Publication Type] OR "Systematic Review"[Publication Type])
The purpose of this search query is to find articles in the PubMed database that have "Curcumin" or "Turmeric" in their titles or abstracts, and these articles are of the types review, meta-analysis, or systematic review. Two field tags are used here:[Title/Abstract]and[Publication Type].[Title/Abstract]indicates that the search terms appear in the title or abstract of the literature, while[Publication Type]is used to specify the type of literature. TheORoperator indicates that the search results can include any of the keywords or conditions. -
("Curcumin"[MeSH Terms] OR "Curcumin"[All Fields]) AND ("Review"[Publication Type])
This search query uses Medical Subject Headings (MeSH Terms) and All Fields to find literature containing "Curcumin" and limits them to the type review. MeSH Terms are professional terms used in PubMed to improve search accuracy. TheANDoperator here indicates that the search results must satisfy all conditions simultaneously. -
("Curcuma longa"[MeSH Terms] OR "Turmeric"[All Fields]) AND ("Curcumin"[All Fields]) AND ("Review"[Publication Type])
This search query adds a search for "Curcuma longa" (the scientific name for turmeric) based on the second search query, ensuring coverage of more literature possibly related to curcumin. At the same time, it also limits the literature type to review. -
("Curcumin"[Title/Abstract] AND "Health Effects"[Title/Abstract]) AND ("Review"[Publication Type])
This search query specifically targets review articles that simultaneously contain "Curcumin" and "Health Effects" in their titles or abstracts. This means that the search results will focus on review articles discussing the health effects of curcumin.
When using these search queries, you can adjust them according to your specific needs. For example, if you are interested in a specific health effect, you can replace "Health Effects" with more specific terms such as "Anti-inflammatory Effects" or "Cancer Prevention". Additionally, you can use parentheses to organize complex search queries and ensure the correctness of the search logic.