I am using Elasticsearch with no modifications whatsoever. This means the mappings, norms, and analyzed/not_analyzed is all default config. I have a very small data set of two items for experimentation purposes. The items have several fields but I query only on one, which is a multi-valued/array of strings field. The doc looks like this:
{
"_index": "index_profile",
"_type": "items",
"_id": "ega",
"_version": 1,
"found": true,
"_source": {
"clicked": [
"ega"
],
"profile_topics": [
"Twitter",
"Entertainment",
"ESPN",
"Comedy",
"University of Rhode Island",
"Humor",
"Basketball",
"Sports",
"Movies",
"SnapChat",
"Celebrities",
"Rite Aid",
"Education",
"Television",
"Country Music",
"Seattle",
"Beer",
"Hip Hop",
"Actors",
"David Cameron",
... // other topics
],
"id": "ega"
}
}
A sample query is:
GET /index_profile/items/_search
{
"size": 10,
"query": {
"bool": {
"should": [{
"terms": {
"profile_topics": [
"Basketball"
]
}
}]
}
}
}
Again there are only two items and the one listed should match the query because the profile_topics field matches with the "Basketball" term. The other item does not match. I only get a result if I ask for clicked = ega in the should.
With Solr I would probably specify that the fields are multi-valued string arrays and are to have no norms and no analyzer so profile_topics are not stemmed or tokenized since all values should be treated as tokens (even the spaces). Not sure this would solve the problem but it is how I treat similar data on Solr.
I assume I have run afoul of some norm/analyzer/TF-IDF issue, if so how do I solve this so that even with two items the query will return ega. If possible I'd like to solve this index or type wide rather than field specific.
Basketball (with capital B) in terms will not be analyzed. This means this is the way it will be searched in the Elasticsearch index.
You say you have the defaults. If so, indexing Basketball under profile_topics field means that the actual term in the index will be basketball (with lowercase b) which is the result of the standard analyzer. So, either you set profile_topics as not_analyzed or you search for basketball and not Basketball.
Read this about terms.
Regarding to setting all the fields to not_analyzed you could do that with a dynamic template. Still with a template you can do what Logstash is doing: defining a .raw subfield for each string field and only this subfield is not_analyzed. The original/parent field still holds the analyzed version of the same text, maybe you will use in the future the analyzed field.
Take a look at this dynamic template. It's the one Logstash is using.
More specifically:
{
"template": "your_indices_name-*",
"mappings": {
"_default_": {
"_all": {
"enabled": true,
"omit_norms": true
},
"dynamic_templates": [
{
"string_fields": {
"match": "*",
"match_mapping_type": "string",
"mapping": {
"type": "string",
"index": "analyzed",
"omit_norms": true,
"fields": {
"raw": {
"type": "string",
"index": "not_analyzed"
}
}
}
}
}
]
}
}
}
Related
I have a simple string "PART_NUMBER" value as a field in solr. I would like to add an additional field which places that value in a URL field. To do this, I created a new field type, field, and copy field
"add-field-type": {
"name": "endpoint_url",
"class": "solr.TextField",
"positionIncrementGap": "100",
"analyzer": {
"tokenizer": {
"class": "solr.KeywordTokenizerFactory"
},
"filters": [
{
"class": "solr.PatternReplaceFilterFactory",
"pattern": "([\\s\\S]*)",
"replacement": "http://myurl/$1.jpg"
}
]
}
},
"add-field": {
"name": "URL",
"type": "endpoint_url",
"stored": true,
"indexed": true
},
"add-copy-field":{ "source":"PART_NUMBER", "dest":"URL" }
As some of you probably guessed, my query output looks like
{
"id": "1",
"PART_NUMBER": "ABCD1234",
"URL": "ABCD1234",
"_version_": 1645658574812086272
}
Because the endpoint_url fieldtype only modifies the index. Indeed, when doing my analysis, I get
http://myurl/ABCD1234.jpg
My question: Is there any way to apply a tokenizer or filter and feed it back in to the field value? I would prefer this output when returning the result:
{
"id": "1",
"PART_NUMBER": "ABCD1234",
"URL": "http://myurl/ABCD1234.jpg",
"_version_": 1645658574812086272
}
Is this possible to do in Solr?
Solution was posted here:
Custom Solr analyzers not being used during indexing
I need to use an Update Processors In order to change the field value before analysis. The process can be found here:
https://lucene.apache.org/solr/guide/8_1/update-request-processors.html
When creating an index definition in Azure Search, is there any way to add additional stop words just for that index. For example if you are indexing street names one would like to strip out Road, Close, Avenue etc.
And if one makes the field non-searchable i.e. the whole thing is indexed as one term, then what happens to something like Birken Court Road. Would the term being indexed be Birken Court.
Many thanks
You can define an additional set of stopwords using a custom analyzer.
For example,
{
"name":"myindex",
"fields":[
{
"name":"id",
"type":"Edm.String",
"key":true,
"searchable":false
},
{
"name":"text",
"type":"Edm.String",
"searchable":true,
"analyzer":"my_analyzer"
}
],
"analyzers":[
{
"name":"my_analyzer",
"#odata.type":"#Microsoft.Azure.Search.CustomAnalyzer",
"tokenizer":"standard_v2",
"tokenFilters":[
"lowercase",
"english_stopwords",
"my_stopwords"
]
}
],
"tokenFilters":[
{
"name":"english_stopwords",
"#odata.type":"#Microsoft.Azure.Search.StopwordsTokenFilter",
"stopwordsList":"english"
},
{
"name":"my_stopwords",
"#odata.type":"#Microsoft.Azure.Search.StopwordsTokenFilter",
"stopwords": ["road", "avenue"]
}
]
}
In this index definition I'm setting a custom analyzer on the text field that used the standard tokenizer, lowercase token filter and two stopwords token filters, one for standard english stopwords and one for the additional set of stopwords. You can test the behavior of your custom analyzer with the Analyze API, for example:
request:
{
"text":"going up the road",
"analyzer": "my_analyzer"
}
response:
{
"tokens": [
{
"token": "going",
"startOffset": 0,
"endOffset": 5,
"position": 0
},
{
"token": "up",
"startOffset": 6,
"endOffset": 8,
"position": 1
}
]
}
Analyzers are not applied to non-searchable fields, therefore the stopword in your example would not be removed. To learn more about query and document processing see: How full text search works in Azure Search.
In my data, I have two fields that I want to use as an index together. They are sensorid (any string) and timestamp (yyyy-mm-dd hh:mm:ss).
So I made an index for these two using the Cloudant index generator. This was created successfully and it appears as a design document.
{
"index": {
"fields": [
{
"name": "sensorid",
"type": "string"
},
{
"name": "timestamp",
"type": "string"
}
]
},
"type": "text"
}
However, when I try to make the following query to find all documents with a timestamp newer than some value, I am told there is no index available for the selector:
{
"selector": {
"timestamp": {
"$gt": "2015-10-13 16:00:00"
}
},
"fields": [
"_id",
"_rev"
],
"sort": [
{
"_id": "asc"
}
]
}
What have I done wrong?
It seems to me like cloudant query only allows sorting on fields that are part of the selector.
Therefore your selector should include the _id field and look like:
"selector":{
"_id":{
"$gt":0
},
"timestamp":{
"$gt":"2015-10-13 16:00:00"
}
}
I hope this works for you!
I'm using Elasticsearch for this project but a Solr solution might be appropriate too. In the query I'd like to include a portion of a should clause that will return results even if none of the other terms can. This will be used for document popularity. I'll periodically calculate reading popularity and add a float field to each doc with a numeric value.
The idea is to return docs based on terms but when that fails, return popular docs ranked by popularity. These should be ordered by term match scores or magnitude of popularity score.
I realize that I could quantize the popularity and treat it like a tag "hottest", "hotter", "hot"... but would like to use numeric field since the ranking is well defined.
Here is the current form of my data (from fetch by id):
GET /index/docs/ipad
returns a sample object
{
"_index": "index",
"_type": "docs",
"_id": "doc1",
"_version": 1,
"found": true,
"_source": {
"category": ["tablets", "electronics"],
"text": ["buy", "an", "ipad"],
"popularity": 0.95347457,
"id": "doc1"
}
}
Current query format
POST /index/docs/_search
{
"size": 10,
"query": {
"bool": {
"should": [
{"terms": {"text": ["ipad"]}}
],
"must": [
{"terms": {"category": ["electronics"]}}
]
}
}
}
This may seem an odd query format but these are structured objects, not free form text.
Can I add popularity to this query so that it returns items ranked by popularity magnitude along with those returned by the should terms? I'd boost the actual terms above the popularity so they'd be favored.
Note I do not want to boost by popularity, I want to return popular if the rest of the query returns nothing.
One approach I can think of is wrapping match_all filter in constant score
and using sort on score followed by popularity
example:
{
"size": 10,
"query": {
"bool": {
"should": [
{
"terms": {
"text": [
"ipad"
]
}
},
{
"constant_score": {
"filter": {
"match_all": {}
},
"boost": 0
}
}
],
"must": [
{
"terms": {
"category": [
"electronics"
]
}
}
],
"minimum_should_match": 1
}
},
"sort": [
{
"_score": {
"order": "desc"
}
},
{
"popularity": {
"unmapped_type": "double"
}
}
]
}
You want to look into the function score query and a decay function for this.
Here's a gentle intro: https://www.found.no/foundation/function-scoring/
Here i quoted my code for multiple search filtering. I could not find the mistakes in that. please give a right code to make it work well.
Employee document:
{
"_id": "527c8d9327c6f27f17df0d2e17000530",
"_rev": "24-276a8dc913559901897fd601d2f9654f",
"proj_role": "TeamMember",
"work_total_experience": "3",
"personal": {
"languages_known": [
"English","Telugu"
]},
"skills": [
{
"skill_set": "Webservices Framework",
"skill_exp": 1,
"skill_certified": "yes",
"skill_rating": 3,
},
{
"skill_set": "Microsoft",
"skill_exp": 1,
"skill_certified": "yes",
"skill_rating": 3,
}
]
"framework_competency": "Nasscom",
"type": "employee-docs"
}
Design Document:
{
"_id": "_design/sample",
"_rev": "86-1250f792e6e84f6f33447a00cf64d61d",
"views": {},
"language": "javascript",
"indexes": {
"search": {
"index": "function(doc){\n index(\"default\", doc._id);if(doc.type=='employee-docs'){\nif (doc.proj_role){index(\"project_role\", doc.proj_role);}if(doc.work_total_experience){\nindex(\"work_experience\", doc.work_total_experience);}\nif(doc.personal.languages_known){for(c in doc.personal.languages_known){ \n index(\"languages_known\",doc.personal.languages_known[c]);}} if(doc.skills){for (var i=0;i<doc.skills.length;i++){\nindex('skill_set',doc.skills[i].skill_set);}}}}"
}
}
}
Run using below URL : https://ideyeah4.cloudant.com/opteamize_new/_design/sample/_search/search?q=project_role:TeamMember%20AND%20work_experience:%223%22%20AND%20languages_known:Telugu%20AND%20skill_set:Microsoft&include_docs=true
A simple way to debug this is to query the top 100 results in your index:
https://ideyeah4.cloudant.com/opteamize_new/_design/sample/_search/search?q=*:*&limit=100
This will at least tell you whether there are any documents in your index at all.
Your current query (without URL encoding) looks like:
project_role:TeamMember AND work_experience:"3" AND languages_known:Telugu AND skill_set:Microsoft
I'd suggest that some of these search values require quotes - always true when you are searching string values. Next, you could try:
project_role:"TeamMember"
see if you get any results and refine from there.
Debugging this might also be easier if you store the values as well as index them (so you can see exactly what is indexed). To do this, add an object to each index call { "store": true }. For example,
index("languages_known", doc.personal.languages_known[c], { "store": true });
Now, when you query the index it will return a list of fields which were stored with each match.