Skip to main content

Record to Record Recommendations for Invenio.

Project description

The Record Recommender creates recommendations for Invenio. By reading page views and downloads from Elasticsearch calculating the recommendations and storing them in Redis to be retrieve and displayed by Invenio.


The workflow to generate the recommendations is:

  1. recommender fetch 24 to cache the last 24 weeks of page views and downloads from Elasticsearch.
  2. recommender profiles 24 generate the user profiles from the page views and downloads. For the not logged in users profiles based on the ip-address and user agent are created.
  3. recommender build 50 calculates the recommendations using 50 processes and stores them in the specified Redis server.

Alternative the recommendations can be automatically be fetched, the profiles generated and the recommendations calculated all this with one command:

recommender update_recommender 24 50 for the last 24 weeks and using 50 processes.


The configuration file is expected to be in /etc/record_recommender.yml otherwise the path to the config file can be defined by using the command line option --config_path.

# Record-Recommender configuration.

es_index: ['index-2014', 'index-2015', 'index-2016']
es_user: user
es_host: localhost
es_port: 443

recommendation_version: 1

# Sentry connection string

host: localhost
port: 6379
db: 0
prefix: 'Reco_1::'

base_path: cache/
cache_file_prefix: ''

version: 1
disable_existing_loggers: False
    format: '%(asctime)s - %(name)s - %(levelname)s - %(message)s'
    class: logging.StreamHandler
    level: INFO
    formatter: simple
    stream: ext://sys.stdout
    class: raven.handlers.logging.SentryHandler
    level: WARN
    class : logging.FileHandler
    formatter: simple
    level: DEBUG
    filename: log_recommender.log
    level: DEBUG
    handlers: [console, file, sentry]
    propagate: no
    # elasticsearch:
    #   level: WARN
    #   handlers: [console]
    #   propagate: no
    level: ERROR
    handlers: [console, file, sentry]

Additional to the configuration options found in the config file environment variables, which overwrite the ones from the config file can be set.

  • RECOMMENDER_ES_PASSWORD to set the Elasticsearch password.
  • RECOMMENDER_SENTRY to set the Sentry connection string.

Command line

Usage: recommender [OPTIONS] COMMAND [ARGS]...

Record-Recommender command line version.

-v, --verbose           Enables verbose mode.
-c, --config_path TEXT  Path to the configuration file.
--help                  Show this message and exit.

debug               Debug the application and recommender.
fetch               Fetch newest PageViews and Downloads.
build               Calculate all recommendations.
profiles            Number of weeks to build.
update_recommender  Download and build the recommendations.

Debugging the Recommendations

As first step look into the created user profiles in the defined cache folder.

To explore the graph with all loaded data use the recommender debug command to get a interactive python shell.

Project details

Release history Release notifications

This version
History Node


History Node


History Node


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
record-recommender-0.0.2.tar.gz (25.6 kB) Copy SHA256 hash SHA256 Source None May 17, 2016

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging CloudAMQP CloudAMQP RabbitMQ AWS AWS Cloud computing Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page