Please input the URL:


Usage Instructions:

  1. URL must be complete. Ex: http://searchjournal.com/, https://searchjournal.com/
  2. URL must be the homepage of journal website.
  3. It is not recommended to use non-journal web URLs.
  4. Don't submit too frequently. The limitation is 15 times per minute.
  5. Click the link below to check out source code:
    6. Github link


Limitations:

  1. Limited knowledge of legitimate and predatory academic journals and publisher websites after January 2021.
  2. The results are for reference only.
  3. May occasionally produce incorrect judgments based on input from different internal pages of the same journal.
  4. The query results may have been inaccurate before the flow control was implemented on the website on March 30, 2023.
  5. The system judgment provides an initial reference, not an absolute answer. Furthermore, interested researchers can continuously optimize the model and source code.

Important Announcement:

Since the publication of this article on February 20, 2023, many interested researchers worldwide have started using the AJPC website to search for journals. However, initially, the website had no search restrictions, and activities such as automatic or mass sending of requests caused the website to crash (see as Fig. 1). Researchers from different countries used it and continue to request other journal URLs. This condition will cause a similar DDOS attack mode. As a result, many researchers concerned with predatory issues have noticed changes in the returned results.
Fig. 1 Users send a large number of requests in a short period of time.

Possible reasons are as follows:
  1. After receiving many requests, the IP address of the AJPC system was banned by journals' websites with protective mechanisms, restricting access to the website's content and returning empty values, leading to a judgment of predatory behavior (see as Fig. 2).
  2. Websites that prevent crawling cannot judge the content.

Fig. 2 Journal URLs with protection mechanisms prohibit the system from accessing the page after a large number of requests.
Since the model has not been changed, the judgment results are based on the same model in November 2020. The problem currently occurs in the I/O process of the server and web pages (see as Fig. 3). Based on users' feedback, as of March 30, 2023, we have made the following changes to the settings:
  1. Limiting the number of queries from the same IP address to 15 per minute. However, if different IP addresses are used to request the same journal website, the protected journal website may still be blocked by the protective mechanism.
  2. Modify the returned results of the uncrawlable webpages to "fail to crawl," and model judgments will be made based on the returned website content.


Fig. 3 Before the network restriction is fixed, the original model uses data that returns an exception or a null value for judgment.
Finally, the model is based on the features generated from 833 blacklists and 1213 whitelists collected before November 11, 2020, providing a preliminary judgment result rather than an absolute answer. The retrieved journal webpages need further analysis if there are significant changes in the judgment results. We have opened the source code on GitHub for interested researchers to improve and optimize. We welcome researchers to collaborate on exploring and improving the model.

Copyright(c) NYCU Distributed Computing System Lab ALL RIGHTS RESERVED.

National Yang Ming Chiao Tung University  |  Department of Computer Science  |  DCS Lab