问题：
弹性搜索MoreLikeThis查询从不返回结果
薛霄
创建新的。NET Framework 4.6.1控制台应用程序
为NEST 6.5.0和ElasticSearch.NET 6.5.0添加NuGet包
然后，我创建了一个新的弹性索引，其中包含具有“tags”属性的对象（类型为“mything”）。此标记是一组可能值中的随机逗号分隔的单词集。在测试中，我在索引中插入了100到5000项。我试了越来越少的可能的词在设置。
var result = EsClient.Search<MyThing>(s => s
.Index(DEFAULT_INDEX)
.Query(esQuery =>
{
var mainQuery = esQuery
.MoreLikeThis(mlt => mlt
.Include(true)
.Fields(f => f.Field(ff => ff.Tags, 5))
.Like(l => l.Document(d => d.Id(id)))
);
return mainQuery;
}
using Nest;
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
namespace Test_MoreLikeThis_ES6
{
class Program
{
public class MyThing
{
public string Tags { get; set; }
}
const string ELASTIC_SERVER = "http://localhost:9200";
const string DEFAULT_INDEX = "my_index";
const int NUM_RECORDS = 1000;
private static Uri es_node = new Uri(ELASTIC_SERVER);
private static ConnectionSettings settings = new ConnectionSettings(es_node).DefaultIndex(DEFAULT_INDEX);
private static ElasticClient EsClient = new ElasticClient(settings);
private static Random rnd = new Random();
static void Main(string[] args)
{
Console.WriteLine("Rebuild index? (y):");
var answer = Console.ReadLine().ToLower();
if (answer == "y")
{
RebuildIndex();
for (int i = 0; i < NUM_RECORDS; i++)
{
AddToIndex();
}
}
Console.WriteLine("");
Console.WriteLine("Getting a Thing...");
var aThingId = GetARandomThingId();
Console.WriteLine("");
Console.WriteLine("Looking for something similar to document with id " + aThingId);
Console.WriteLine("");
Console.WriteLine("");
GetMoreLikeAThing(aThingId);
}
private static string GetARandomThingId()
{
var firstdocQuery = EsClient
.Search<MyThing>(s =>
s.Size(1)
.Query(q => {
return q.FunctionScore(fs => fs.Functions(fn => fn.RandomScore(rs => rs.Seed(DateTime.Now.Ticks).Field("_seq_no"))));
})
);
if (!firstdocQuery.IsValid || firstdocQuery.Hits.Count == 0) return null;
var hit = firstdocQuery.Hits.First();
Console.WriteLine("Found a thing with id '" + hit.Id + "' and tags: " + hit.Source.Tags);
return hit.Id;
}
private static void GetMoreLikeAThing(string id)
{
var result = EsClient.Search<MyThing>(s => s
.Index(DEFAULT_INDEX)
.Query(esQuery =>
{
var mainQuery = esQuery
.MoreLikeThis(mlt => mlt
.Include(true)
.Fields(f => f.Field(ff => ff.Tags, 5))
.Like(l => l.Document(d => d.Id(id)))
);
return mainQuery;
}
));
if (result.IsValid)
{
if (result.Hits.Count > 0)
{
Console.WriteLine("These things are similar:");
foreach (var hit in result.Hits)
{
Console.WriteLine(" " + hit.Id + " : " + hit.Source.Tags);
}
}
else
{
Console.WriteLine("No similar things found.");
}
}
else
{
Console.WriteLine("There was an error running the ES query.");
}
Console.WriteLine("");
Console.WriteLine("Enter (y) to get another thing, or anything else to exit");
var y = Console.ReadLine().ToLower();
if (y == "y")
{
var aThingId = GetARandomThingId();
GetMoreLikeAThing(aThingId);
}
Console.WriteLine("");
Console.WriteLine("Any key to exit...");
Console.ReadKey();
}
private static void RebuildIndex()
{
var existsResponse = EsClient.IndexExists(DEFAULT_INDEX);
if (existsResponse.Exists) //delete existing mapping (and data)
{
EsClient.DeleteIndex(DEFAULT_INDEX);
}
var rebuildResponse = EsClient.CreateIndex(DEFAULT_INDEX, c => c.Settings(s => s.NumberOfReplicas(1).NumberOfShards(5)));
var response2 = EsClient.Map<MyThing>(m => m.AutoMap());
}
private static void AddToIndex()
{
var myThing = new MyThing();
var tags = new List<string> {
"catfish",
"tractor",
"racecar",
"airplane",
"chicken",
"goat",
"pig",
"horse",
"goose",
"duck"
};
var randNum = rnd.Next(0, tags.Count);
//get randNum random tags
var rand = tags.OrderBy(o => Guid.NewGuid().ToString()).Take(randNum);
myThing.Tags = string.Join(", ", rand);
var ir = new IndexRequest<MyThing>(myThing);
var indexResponse = EsClient.Index(ir);
Console.WriteLine("Index response: " + indexResponse.Id + " : " + string.Join(" " , myThing.Tags));
}
}
}
共有1个答案
鲍俊杰
这里的问题是，对于原型文档的任何术语，默认值min_term_freq永远不会满足2，因为所有文档只包含每个标记（术语）一次。如果将min_term_freq降为1，就会得到结果。还可能需要将min_doc_freq设置为1，并与排除原型文档的查询组合。
这里有一个可以玩的例子
const string ELASTIC_SERVER = "http://localhost:9200";
const string DEFAULT_INDEX = "my_index";
const int NUM_RECORDS = 1000;
private static readonly Random _random = new Random();
private static readonly IReadOnlyList<string> Tags =
new List<string>
{
"catfish",
"tractor",
"racecar",
"airplane",
"chicken",
"goat",
"pig",
"horse",
"goose",
"duck"
};
private static ElasticClient _client;
private static void Main()
{
var pool = new SingleNodeConnectionPool(new Uri(ELASTIC_SERVER));
var settings = new ConnectionSettings(pool)
.DefaultIndex(DEFAULT_INDEX);
_client = new ElasticClient(settings);
Console.WriteLine("Rebuild index? (y):");
var answer = Console.ReadLine().ToLower();
if (answer == "y")
{
RebuildIndex();
AddToIndex();
}
Console.WriteLine();
Console.WriteLine("Getting a Thing...");
var aThingId = GetARandomThingId();
Console.WriteLine();
Console.WriteLine("Looking for something similar to document with id " + aThingId);
Console.WriteLine();
Console.WriteLine();
GetMoreLikeAThing(aThingId);
}
public class MyThing
{
public List<string> Tags { get; set; }
}
private static string GetARandomThingId()
{
var firstdocQuery = _client
.Search<MyThing>(s =>
s.Size(1)
.Query(q => q
.FunctionScore(fs => fs
.Functions(fn => fn
.RandomScore(rs => rs
.Seed(DateTime.Now.Ticks)
.Field("_seq_no")
)
)
)
)
);
if (!firstdocQuery.IsValid || firstdocQuery.Hits.Count == 0) return null;
var hit = firstdocQuery.Hits.First();
Console.WriteLine($"Found a thing with id '{hit.Id}' and tags: {string.Join(", ", hit.Source.Tags)}");
return hit.Id;
}
private static void GetMoreLikeAThing(string id)
{
var result = _client.Search<MyThing>(s => s
.Index(DEFAULT_INDEX)
.Query(esQuery => esQuery
.MoreLikeThis(mlt => mlt
.Include(true)
.Fields(f => f.Field(ff => ff.Tags))
.Like(l => l.Document(d => d.Id(id)))
.MinTermFrequency(1)
.MinDocumentFrequency(1)
) && !esQuery
.Ids(ids => ids
.Values(id)
)
)
);
if (result.IsValid)
{
if (result.Hits.Count > 0)
{
Console.WriteLine("These things are similar:");
foreach (var hit in result.Hits)
{
Console.WriteLine($" {hit.Id}: {string.Join(", ", hit.Source.Tags)}");
}
}
else
{
Console.WriteLine("No similar things found.");
}
}
else
{
Console.WriteLine("There was an error running the ES query.");
}
Console.WriteLine();
Console.WriteLine("Enter (y) to get another thing, or anything else to exit");
var y = Console.ReadLine().ToLower();
if (y == "y")
{
var aThingId = GetARandomThingId();
GetMoreLikeAThing(aThingId);
}
Console.WriteLine();
Console.WriteLine("Any key to exit...");
}
private static void RebuildIndex()
{
var existsResponse = _client.IndexExists(DEFAULT_INDEX);
if (existsResponse.Exists) //delete existing mapping (and data)
{
_client.DeleteIndex(DEFAULT_INDEX);
}
var rebuildResponse = _client.CreateIndex(DEFAULT_INDEX, c => c
.Settings(s => s
.NumberOfShards(1)
)
.Mappings(m => m
.Map<MyThing>(mm => mm.AutoMap())
)
);
}
private static void AddToIndex()
{
var bulkAllObservable = _client.BulkAll(GetMyThings(), b => b
.RefreshOnCompleted()
.Size(1000));
var waitHandle = new ManualResetEvent(false);
Exception exception = null;
var bulkAllObserver = new BulkAllObserver(
onNext: r =>
{
Console.WriteLine($"Indexed page {r.Page}");
},
onError: e =>
{
exception = e;
waitHandle.Set();
},
onCompleted: () => waitHandle.Set());
bulkAllObservable.Subscribe(bulkAllObserver);
waitHandle.WaitOne();
if (exception != null)
{
throw exception;
}
}
private static IEnumerable<MyThing> GetMyThings()
{
for (int i = 0; i < NUM_RECORDS; i++)
{
var randomTags = Tags.OrderBy(o => Guid.NewGuid().ToString())
.Take(_random.Next(0, Tags.Count))
.OrderBy(t => t)
.ToList();
yield return new MyThing { Tags = randomTags };
}
}
下面是一个输出示例
Found a thing with id 'Ugg9LGkBPK3n91HQD1d5' and tags: airplane, goat
These things are similar:
4wg9LGkBPK3n91HQD1l5: airplane, goat
9Ag9LGkBPK3n91HQD1l5: airplane, goat
Vgg9LGkBPK3n91HQD1d5: airplane, goat, goose
sQg9LGkBPK3n91HQD1d5: airplane, duck, goat
lQg9LGkBPK3n91HQD1h5: airplane, catfish, goat
9gg9LGkBPK3n91HQD1l5: airplane, catfish, goat
FQg9LGkBPK3n91HQD1p5: airplane, goat, goose
Jwg9LGkBPK3n91HQD1p5: airplane, goat, goose
Fwg9LGkBPK3n91HQD1d5: airplane, duck, goat, tractor
Kwg9LGkBPK3n91HQD1d5: airplane, goat, goose, horse
类似资料：
-
我正在LDAP服务器上工作。它有弹性搜索。我必须用一些Javascript代码（JSON格式）发送查询。 这是我的查询： 我试图打印所有结果，其中“server”=“server\u name”（该字段是server:server\u name…）。我认为关于弹性搜索的文档太小了。我找到了一些文档，但都是一样的，对新用户没有帮助。这个例子太简单了。 此查询返回所有结果，包括任何筛选器。 Ps：这就
-
在我的弹性搜索数据库中插入6条记录后，我正在尝试一个弹性搜索示例。
-
我有以下格式的弹性搜索文档 } } 我的要求是，当我搜索特定字符串（string.string）时，我只想获得该字符串的FileOffSet（string.FileOffSet）。我该怎么做？ 谢谢
-
试图找出这个微不足道的例子的分数。我希望得到brenda eaton的文件，但我得到的是brenda fassie的最佳结果。
-
以下是我的疑问.... 我没有结果。 另外，我正在使用这个插件来生成请求正文。 我的查询如下所示.. null 感谢您到目前为止的阅读，如果有人能帮助我找出如何使这一工作，我将非常感谢。
-
我有一个问题，我想把弹性搜索的结果减少到1000个，不管有多少匹配的结果匹配，但这不应该影响排名和得分。 我在尝试，但这似乎只是告诉弹性搜索只获得前N个结果，而不考虑分数。如果我说错了，请纠正我。 有什么帮助吗？ 编辑： 我已经在使用分页。因此，在From/Size中使用Size只会影响当前页面的大小。但我想将总结果的大小限制为1000，然后对其进行分页。