[Semantic search]({{site.url}}{{site.baseurl}}/search-plugins/semantic-search/) relies on dense retrieval that is based on text embedding models. However, dense methods use k-NN search, which consumes a large amount of memory and CPU resources. An alternative to semantic search, neural sparse search is implemented using an inverted index and is thus as efficient as BM25. Neural sparse search is facilitated by sparse embedding models. When you perform a neural sparse search, it creates a sparse vector (a list of `token: weight` key-value pairs representing an entry and its weight) and ingests data into a rank features index.
Before using neural sparse search, make sure to set up a [pretrained sparse embedding model]({{site.url}}{{site.baseurl}}/ml-commons-plugin/pretrained-models/#sparse-encoding-models) or your own sparse embedding model. For more information, see [Choosing a model]({{site.url}}{{site.baseurl}}/ml-commons-plugin/integrating-ml-models/#choosing-a-model).
1. [Create an ingest pipeline](#step-1-create-an-ingest-pipeline).
1. [Create an index for ingestion](#step-2-create-an-index-for-ingestion).
1. [Ingest documents into the index](#step-3-ingest-documents-into-the-index).
1. [Search the index using neural search](#step-4-search-the-index-using-neural-search).
## Step 1: Create an ingest pipeline
To generate vector embeddings, you need to create an [ingest pipeline]({{site.url}}{{site.baseurl}}/api-reference/ingest-apis/index/) that contains a [`sparse_encoding` processor]({{site.url}}{{site.baseurl}}/api-reference/ingest-apis/processors/sparse-encoding/), which will convert the text in a document field to vector embeddings. The processor's `field_map` determines the input fields from which to generate vector embeddings and the output fields in which to store the embeddings.
The following example request creates an ingest pipeline where the text from `passage_text` will be converted into text embeddings and the embeddings will be stored in `passage_embedding`:
In order to use the text embedding processor defined in your pipeline, create a rank features index, adding the pipeline created in the previous step as the default pipeline. Ensure that the fields defined in the `field_map` are mapped as correct types. Continuing with the example, the `passage_embedding` field must be mapped as [`rank_features`]({{site.url}}{{site.baseurl}}/field-types/supported-field-types/rank/#rank-features). Similarly, the `passage_text` field should be mapped as `text`.
The following example request creates a rank features index that is set up with a default ingest pipeline:
To save disk space, you can exclude the embedding vector from the source as follows:
```json
PUT /my-nlp-index
{
"settings": {
"default_pipeline": "nlp-ingest-pipeline-sparse"
},
"mappings": {
"_source": {
"excludes": [
"passage_embedding"
]
},
"properties": {
"id": {
"type": "text"
},
"passage_embedding": {
"type": "rank_features"
},
"passage_text": {
"type": "text"
}
}
}
}
```
{% include copy-curl.html %}
Once the `<token, weight>` pairs are excluded from the source, they cannot be recovered. Before applying this optimization, make sure you don't need the `<token, weight>` pairs for your application.
To ingest documents into the index created in the previous step, send the following requests:
```json
PUT /my-nlp-index/_doc/1
{
"passage_text": "Hello world",
"id": "s1"
}
```
{% include copy-curl.html %}
```json
PUT /my-nlp-index/_doc/2
{
"passage_text": "Hi planet",
"id": "s2"
}
```
{% include copy-curl.html %}
Before the document is ingested into the index, the ingest pipeline runs the `sparse_encoding` processor on the document, generating vector embeddings for the `passage_text` field. The indexed document includes the `passage_text` field, which contains the original text, and the `passage_embedding` field, which contains the vector embeddings.
To perform a neural sparse search on your index, use the `neural_sparse` query clause in [Query DSL]({{site.url}}{{site.baseurl}}/opensearch/query-dsl/index/) queries.
