bala n
Reputation: 39
ElasticSearch is not returning the document in correct order
For the below query, I am searching the text "christ". the result is not showing the correct order. I have a record for "Christine" but it is not showing as a first document. Can you help me to fix this order issue? I am using Elasticsearch 8.5.3. but the same is correctly working in Elasticsearch 2.x version.
Mapping and Settings are below,
{
"template": {
"settings": {
"index": {
"max_ngram_diff": "3",
"routing": {
"allocation": {
"include": {
"_tier_preference": "data_content"
}
}
},
"analysis": {
"analyzer": {
"special_char_edge_1_gram_analyzer": {
"filter": [
"lowercase"
],
"char_filter": [
"string_special_character_char_filter"
],
"type": "custom",
"tokenizer": "edge_1_gram_tokenizer"
},
"tokenizer": {
"edge_1_gram_tokenizer": {
"token_chars": [
"letter",
"digit"
],
"min_gram": "1",
"type": "edge_ngram",
"max_gram": "100"
}
},
"number_of_shards": "1",
"number_of_replicas": "1"
}
},
"mappings": {
"name": {
"properties": {
"first": {
"type": "text",
"fields": {
"special_char_edge_1_gram_analyzer": {
"type": "text",
"analyzer": "special_char_edge_1_gram_analyzer"
}
}
}
}
}
},
"aliases": {}
}
}
}
}
}
}
}
Query is below,
GET as.contact-154/_search
{
"query": {
"bool": {
"should": {
"multi_match": {
"type": "best_fields",
"query": "christ",
"fields": [ "name.first^2", "name.first.special_char_edge_1_gram_analyzer^5" ]
}
}
}
},
"_source": ["id","name.first","name.last"],
"highlight": {
"pre_tags": [
"<span class='text-highlight'>"
],
"post_tags": [
"</span>"
],
"fields": {
"name.first": {},
"id": {}
}
},
"from": 0,
"size": 100
}
Result,
{
"took": 1,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 125,
"relation": "eq"
},
"max_score": 62.842735,
"hits": [
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "127848",
"_score": 62.842735,
"_source": {
"name": {
"last": "Poeschl",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "128911",
"_score": 62.842735,
"_source": {
"name": {
"last": "Rhode",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "128925",
"_score": 62.842735,
"_source": {
"name": {
"last": "Hanson",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "129102",
"_score": 62.842735,
"_source": {
"name": {
"last": "Cooper",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "129116",
"_score": 62.842735,
"_source": {
"name": {
"last": "Sipos",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "129119",
"_score": 62.842735,
"_source": {
"name": {
"last": "Sipos",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "129147",
"_score": 62.842735,
"_source": {
"name": {
"last": "Dudlery",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "1264420",
"_score": 62.842735,
"_source": {
"name": {
"last": "Jaquess",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "1264421",
"_score": 62.842735,
"_source": {
"name": {
"last": "Jaquess",
"first": "Chris"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Chris</span>"
]
}
},
{
"_index": "es.contact-154.23-05-12-7-04-12",
"_id": "127930",
"_score": 60.612587,
"_source": {
"name": {
"last": "Vanderzanden",
"first": "Christine"
}
},
"highlight": {
"name.first.special_char_edge_1_gram_analyzer": [
"<span class='text-highlight'>Christ</span>ine"
]
}
}
]
}
}
Latest Result,
{
"took": 1,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 125,
"relation": "eq"
},
"max_score": 104.737885,
"hits": [
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "127848",
"_score": 104.737885,
"_source": {
"name": {
"last": "Poeschl",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "128911",
"_score": 104.737885,
"_source": {
"name": {
"last": "Rhode",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "128925",
"_score": 104.737885,
"_source": {
"name": {
"last": "Hanson",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "129102",
"_score": 104.737885,
"_source": {
"name": {
"last": "Cooper",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "129116",
"_score": 104.737885,
"_source": {
"name": {
"last": "Sipos",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "129119",
"_score": 104.737885,
"_source": {
"name": {
"last": "Sipos",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "129147",
"_score": 104.737885,
"_source": {
"name": {
"last": "Dudlery",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "1264420",
"_score": 104.737885,
"_source": {
"name": {
"last": "Jaquess",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "1264421",
"_score": 104.737885,
"_source": {
"name": {
"last": "Jaquess",
"first": "Chris"
}
}
},
{
"_index": "es.contact-154.23-05-12-14-34-46",
"_id": "127930",
"_score": 102.020966,
"_source": {
"name": {
"last": "Vanderzanden",
"first": "Christine"
}
}
}
]
}
}
Upvotes: 0
Views: 128
Answers (1)
Val
Reputation: 217464
The problem is that the edge-ngram analyzer is also kicking in at search time, and hence, the searched string christ is resolved to chris by the analyzer at search time and matches the first records. That's not desired.
You should modified your mapping like this and specify a standard search_analyzer so that the searched text doesn't get edge-ngramed:
{
"mappings": {
"name": {
"properties": {
"first": {
"type": "text",
"fields": {
"special_char_edge_1_gram_analyzer": {
"type": "text",
"analyzer": "special_char_edge_1_gram_analyzer",
"search_analyzer": "standard"
}
}
}
}
}
}
}
Upvotes: 0
Related Questions
- Using next js 13 with elastic search
- Elastic search aggregation using min_doc_count=0 returns all the buckets which are not related to query results or hits
- elasticsearch range filter incorrect
- How to return actual value (not lowercase) when performing search with terms aggregation?
- How to createID and pick latest n entries from the set in elasticsearch
- Getting the Double values instead of Integer using JestClient to retrieve document from elasticsearch
- rails : type search with data address use elasticsearch chewy gem
- Configuration a river for MongoDb collection indexed from ES
- How to retrieve the ratings/number of categories with Tire/ElasticSearch?
- ElasticSearch doubles my _id field when filtering on it