"weak relevance" is a correct and usable phrase in written English.
Generally, it is used when discussing the degree of correlation or relations between two topics. For example, "The study found a weak relevance between the two variables."
Thus, while searching for strongly relevant feature-pairs, not only strongly relevant features can be selected, but also putative weakly relevant features can be included, i.e., features of weak relevance are sometimes able to contribute to the classification performance when combined with other features.
Items with weak relevance can then be reworded or excluded from the questionnaire.
After the extension, each protein is associated with a group of GO terms with strong or weak relevance.
Weak relevance suggests that the feature is not always necessary but may become necessary for an optimal subset at certain conditions.
Kohavi and John (hereafter KJ) [ 41] addressed the issue of finding a good feature subset and its relation to which features shall be included by partitioning features into three types of strong relevance, weak relevance, and irrelevance.
A study by Lavis found that the research-to-policy gaps were related to a myriad of factors that policymakers have to consider, such as the value attached to research evidence in decision making, and to weak relevance of the research evidence and the concerns of policy makers at a point in time [ 8].
Accordingly, the gene with the weakest relevance is the last gene in a group of marker genes (the gene associated with g = G).
Therefore, we have to consider the possibility that such SNPs, despite having reached statistical significance in large GWASs, could have a weak biological relevance.
For example, Jackson and Pettit (1988, 1990) distinguish the robust "causal efficacy" of physical properties from the weaker "causal relevance" of higher-level properties.
We argue that query is a weak indicator of relevance in that query is usually much shorter than the relevant documents, such that the use of semantic associations of the query-document pairs may only lead to limited improvement in retrieval performance.
As described in Sect. 1, patient records are much shorter than full-text biomedical articles, such that patient record is a weak indicator of relevance, thus our feedback-based CDS method is expected to outperform the method (BM25+Sim_{d_{Para}text Q}).
