dimitris.pierrakos
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

[cleaning wf] fixed regex used to spot garbage in result titles; adjusted threshold for filtering titles

This commit is contained in:
Claudio Atzori 2021-11-16 15:24:23 +01:00
parent 0a727d325d
commit 49f897ef29
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
dhp-common/src/main/java/eu/dnetlib/dhp/schema/oaf/utils/GraphCleaningFunctions.java
View File

@ -27,8 +27,8 @@ public class GraphCleaningFunctions extends CleaningFunctions {
public static final int ORCID_LEN = 19; public static final int ORCID_LEN = 19;
public static final String CLEANING_REGEX = "(?:\\n|\\r|\\t)"; public static final String CLEANING_REGEX = "(?:\\n|\\r|\\t)";
public static final String INVALID_AUTHOR_REGEX = ".*deactivated.*"; public static final String INVALID_AUTHOR_REGEX = ".*deactivated.*";
public static final String TITLE_FILTER_REGEX = "[.*test.*\\W\\d]"; public static final String TITLE_FILTER_REGEX = "(test)|\\W|\\d";
public static final int TITLE_FILTER_RESIDUAL_LENGTH = 10; public static final int TITLE_FILTER_RESIDUAL_LENGTH = 5;
public static <T extends Oaf> T fixVocabularyNames(T value) { public static <T extends Oaf> T fixVocabularyNames(T value) {
if (value instanceof Datasource) { if (value instanceof Datasource) {