Microsoft Academic Search API dataset in Windows Azure Marketplace RRS feed

  • General discussion

  • Access to the Microsoft Academic Search API is now available through the Windows Azure Marketplace and no longer requires an AppID. To access the dataset, go to the Microsoft Academic Data page, and sign up.

    Thomas, Academic Search Editor

    Wednesday, November 27, 2013 7:08 PM

All replies

  • Is the use of the Windows Azure Marketplace mandatory? Will the old API (JSON/SOAP) and the main Microsoft Academic Search website be turned off?

    Furthermore, will the data set at the Windows Azure Marketplace be held up-to-date? Or is the data set fixed until some specific date (December 2013)?

    I also tried to use the data in the Azure Explorer, but the filter possibilities are very limited. I am only able to hide the rows or to filter the numeric rows by some simple methods. Are there any functions to query the data set or to use some string operations? The old API provided the keyword "FullTextQuery" and I could pass in a parameter to be searched for. The new data set at the Windows Azure Marketplace does not provide any search methods.

    Thank you in advice!

    Saturday, December 28, 2013 4:53 PM
  • Hello Vilius,

    MAS dataset has been made available via Windows Azure datamarket ''https://datamarket.azure.com/dataset/mrc/microsoftacademic''and can be accessed by subscribing to the "Microsoft Academic" dataset. This dataset is receiving fresh papers as they are released on arXiv.org.

    Writing a custom application around this dataset will provide more flexibility. You can know more about query types via the link ''http://go.microsoft.com/fwlink/?LinkID=201299''

    If you are using Visual Studio, then you could refer to link ''http://msdn.microsoft.com/en-us/library/gg312152.aspx''

    Thanks and Regards,
    Rahul Sharma

    Tuesday, January 7, 2014 7:55 PM
  • Hello Rahul,

    thank you for your post and the referenced links. I must admit, that I still have big problems migrating from the old API to the new API (Azure Marketplace). This depends mainly on the poor documentation provided at the Azure Marketplace website. Therefore, I hope that you (or someone else) could answer my following questions.

    1. I encounter many problems trying to combine the possible parameters (https://datamarket.azure.com/dataset/mrc/microsoftacademic#schema) with the root URL (https://api.datamarket.azure.com/MRC/MicrosoftAcademic/v1/). For example, I ran the following command to get the publications from year 2013: https://api.datamarket.azure.com/MRC/MicrosoftAcademic/v1/Paper?Year=2013. The response to my query was "The query parameter 'Year' is not supported" - but it is listed in the table for the possible parameters? So how should these elements be combined?

    2. Hereby, I also refer to the possible parameters for the MAS dataset. As far as I can see, for the publication entities, I am only able to query the ID, Year, ConfID and JourID. But there is no possibility to query for the title? And not only is the search by title not available, but also the fuzzy search method as it is well known from every "conventional" search engine. I mean, it is impossible to search for publications that refer to a search term. Therefore, I am quite disappointed of the Azure Marketplace or rather the MAS dataset. Or am I missing something? Because otherwise, for my purposes, that dataset is useless.

    Thank you in advice,

    Thursday, January 16, 2014 10:20 AM
  • Thanks Vilius for your reply. Given below is answer to your queries:

    1) Odata query for listing papers by year 2013 is ''https://api.datamarket.azure.com/MRC/MicrosoftAcademic/v1/Paper?$filter=Year eq 2013'' . OData query format is bit different and supported query options are given at the link  ''http://msdn.microsoft.com/en-us/library/gg312156.aspx''. Adding another link which has examples: ''http://www.odata.org/documentation/odata-v2-documentation/uri-conventions/''

    2) I agree that not being able to search by paper title or fuzzy search is a drawback. It could not be enabled due a technical limitation. Hopefully we can overcome the limitation and can enable it in future.

    We'll also update the support document and add few query examples. Thanks once again for your valued feedback and support.


    Friday, January 17, 2014 10:11 PM
  • Hello Rahul,

    I am very thankful for your help! Now, I have an overview of the available possibilities. 

    During the evaluation of the new API at the Azure Marketplace, I could not find the maximum amount of allowed API calls per time unit? The old API had a limit of 200 calls per minute, but I could not find any specifications neither on the website (http://datamarket.azure.com/dataset/mrc/microsoftacademic) nor in the schema reference (http://sdrv.ms/1bdJBgD). Therefore, what are the API call limits?

    Thank you in advice,

    Friday, January 24, 2014 1:30 PM
  • Welcome Vilius.

    In this case there seems to be a limit of 300 requests per minute. Advisable to keep it less than 250.


    Saturday, January 25, 2014 2:00 AM
  • Dear Rahul,

    refering to your earlier answer back in January :

    > 2) I agree that not being able to search by paper title or fuzzy search is a drawback. It could not be enabled due a technical limitation. Hopefully we can overcome the limitation and can enable it in future.

    Do you now have news regarding the future availability of a fuzzy search feature ?



    Monday, November 24, 2014 6:36 AM
  • Hello, 

    is there someone providing support on MAS still ?



    Sunday, December 14, 2014 8:15 PM
  • Hi Pierre,

    There is no new news at the moment.

    Best Regards,


    Thomas, Academic Search Editor

    Monday, December 15, 2014 4:32 PM
  • Yes.


    Thomas, Academic Search Editor

    Monday, December 15, 2014 4:33 PM
  • All of the links in this thread that point to MSDN resources return "This topic is no longer available." Can you update those links for us with versions that point to the referenced material? Thanks.
    Tuesday, February 10, 2015 8:13 PM