We have an Azure Search Service that uses Full Lucene.
Our index contains the following data
{
"@odata.context":
"https://inkredible-search.search.windows.net/indexes('index-4587694857649234')/$metadata#docs(*)",
"value": [
{
"@search.score":
1.9101982,
"Id":
"30857",
"Name":
"Brother LC103"
},
{
"@search.score":
1.9101982,
"Id":
"30848",
"Name":
"Brother LC203"
},
{
"@search.score":
1.9101982,
"Id":
"30846",
"Name":
"Brother LC3017"
},
{
"@search.score":
1.9101982,
"Id":
"30845",
"Name":
"Brother LC51"
},
{
"@search.score":
1.9101982,
"Id":
"30844",
"Name":
"Brother LC65"
}
]
}
When searching for LC65 we would expect the item with Name "Brother LC65" to score the highest. It does not.
The same search results are shown when searching for LC, LC65, LC443523
It seems the numeric portion of the string is completely ignored.
Any suggestions for solving this puzzle?
