Zotero MAS MetaData

⚠️ The MAS API is depreciated as of the start of 2022. I am building a new plugin that uses the Semantic Scholar API instead and adds more metadata.

MAS MetaData is a Zotero plugin that uses the Microsoft Academic Search API (MAS) to add more metadata to Zotero. Currently it only adds citation counts to the extra field of items. This will be extended upon in future versions.

You need to generate an api key for the Microsoft Academic Search (MAS) API to use this plugin.

MAS MetaData is based in part on Zotero Google Scholar Citations and Zotero DOI Manager.

Installation

1. Download the latest version of Zotero MAS MetaData
2. Start Zotero and go to Tools -> Add-ons -> Tools for all Add-ons (wheel in the top right corner) -> Install Add-on From File select the downloaded .xpi file and restart
3. Generate an api key at https://msr-apis.portal.azure-api.net/products/project-academic-knowledge
4. Add the key (primary or secondary) to Tools -> MASMetaData Preferences... -> MAS API Key

Academic Search API

Rate Limits

The free api has a rate limit of 10,000 transactions per month, 3 per second for interpret, 1 per second for evaluate, 6 per minute for calcHistogram.

The plugin currently uses at most one interpret request and two evaluate request per item. Thus you can update metadata for 3,333 items per month (per subscription) in the worst case. However the average should be around 9,000 items per month as most items should use only one evaluate request and no interpret requests.

Citation Count

The plugin adds citation counts to the extra field of items in Zotero. The citation count used is the estimated citation count (ECC) from Microsoft, which is calculated using the Microsoft Academic Graph data to get a more accurate citation count for each publication (at least thats what Microsoft says https://academic.microsoft.com/faq?target=ranking1).

Logprob

Successfully query responses from the MAS API come with a probability (prob) value between 0 and 1. This value determines how likely the response is to be correct (higher - more likely to be correct).

In the setting Tools -> MASMetaData Preferences... -> Advanced Settings you can edit a cutoff probability. Responses with lower probability than the cutoff are excluded. The setting uses a logarithm of the probability (logprob) for ease of use.

Further Information

The MAS API is provided by the Project Academic Knowledge from Microsoft.
Docs:
https://msr-apis.portal.azure-api.net/docs/services/academic-search-api/
https://docs.microsoft.com/en-us/academic-services/project-academic-knowledge/

License

The source code is released under the MIT License.