Fusion 5.10
Legacy Product
Train a collaborative filtering matrix decomposition recommender using SparkML’s Alternating Least Squares (ALS) to batch-compute query-query similarities. This can be used for items-for-query recommendations as well as queries-for-query recommendations.
Instances of session in this topic refer to user session.
| This job is deprecated in Fusion 5.1 and removed in 5.2. Use the Query-to-Query Session-Based Similarity Jobs for better performance and query coverage. |
Use this job to to batch compute query-query similarities using a co-occurrence based approach
Solr field containing number of events (e.g., number of clicks).
Default: count_i
Spark-compatible format which training data comes in (like 'solr', 'hdfs', 'file', 'parquet' etc)
Default: solr
Allowed values: solrhdfsfileparquet
Solr field containing document id that user clicked.
Default: doc_id_s
Field containing queries.
>= 1 characters
Default: query_s
The ID for this Spark job. Used in the API to reference this job. Allowed characters: a-z, A-Z, dash (-) and underscore (_). Maximum length: 63 characters.
<= 63 characters
Match pattern: [a-zA-Z][_\-a-zA-Z0-9]*[a-zA-Z0-9]?
Queries above this length will not be considered for generating recommendations.
>= 1
exclusiveMinimum: false
Default: 50
Minimum number of times a query pair must be generated to be considered valid.
>= 1
exclusiveMinimum: false
Default: 2
The minimum number of clicked documents needed for comparing queries.
>= 1
exclusiveMinimum: false
Default: 1
Queries below this length (in number of characters) will not be considered for generating recommendations.
>= 1
exclusiveMinimum: false
Default: 3
Collection to store synonym and similar query pairs.
Maximize score for query pairs with overlapping tokens by setting score to 1.
Default: true
The threshold above which query pairs are consider similar. Decreasing the value can fetch more pairs at the expense of quality.
<= 1
exclusiveMaximum: false
Default: 0.3
For any deterministic pseudorandom number generation
Default: 1234
If session id is not available, specify user id field instead. If this field is left blank, session based recommendations will be disabled.
Default: session_id_s
Spark configuration settings.
object attributes:{key required : {
display name: Parameter Name
type: string
}value : {
display name: Parameter Value
type: string
}}
String of special characters to be filtered from queries.
Default: ~!@#$^%&*\(\)_+={}\[\]|;:"'<,>.?`/\\-
Name of the stopwords blob resource. This is a .txt file with one stopword per line. By default the file is called stopwords/stopwords_nltk_en.txt however a custom file can also be used. Check documentation for more details on format and uploading to blob store.
Default: stopwords/stopwords_nltk_en.txt
Minimum amount of overlap to consider for boosting. To specify overlap in terms of ratio, specify a value in (0, 1). To specify overlap in terms of exact count, specify a value >= 1. If value is 0, boost will be applied if one query is a substring of its pair.Stopwords are ignored while counting overlaps.
Default: 1
Collection containing queries, document id and event counts. Can be either signal aggregation collection or raw signals collection.
Solr query to additionally filter the input collection.
>= 3 characters
Default: *:*
Additional spark dataframe loading configuration options
Fraction of the training data to use
<= 1
exclusiveMaximum: false
Default: 1
Default: similar_queries
Allowed values: similar_queries
Options used when writing output to Solr.
object attributes:{key required : {
display name: Parameter Name
type: string
}value : {
display name: Parameter Value
type: string
}}