Why are my citations and h-index lower than those of Google Scholar?

Google Scholar counts the citations from various types of documents, including web resources. We only count the citations from peer-reviewed articles (plus some book chapters). On the other hand, Google Scholar has a relaxed mechanism for matching publications by author names. As a result, the number of publications of each other is usually overestimated in Google Scholar and underestimated in our system because of a strict matching mechanism (e.g., ORCID, name and institution/department, name and co-authors, etc.).

Not all my publications are in my profile. Can I add them manually?

Unfortunately, no! We do our best to recognise all authors, but it is not always easy. On the other hand, since we calculate the percentiles, it is fair to all authors to let the same system recognise the publications. If some of your publications are missing, it can be the case for others too. Therefore, the percentiles are statistically more reliable.

The core idea is not to be biased toward authors who are active in maintaining their profiles.

I don't like my profile. Can I delete it?

Unfortunately, no! It is not a personal profile but an overview of the scholarly literature. Removing any part of it damage the integrity of the whole system. As an analogue, a public figure cannot request to be removed from encyclopedias.

Search Engine

Why aren't excerpts containing the keywords shown along with the results?

The purpose of the exaly search engine is to find papers. We use algorithms to weight keywords according to their distribution in each paper. Therefore, the finding is not based on a specific part of the article to be highlighted.

What are the meanings of numbers alongside the results found?

In most cases, if you hover over the corresposing text, a brief description appears.

Why is the number of results found for a keyword lower than other search engines?

The first reason is that we only cover peer-reviewed articles or similar documents. The second one is that our search engine is not simply based on the appearance of a keyword in the text. We use NLP to improve search performance. Therefore, some results are excluded by the algorithm if they are not relevant to the topic searched, though they may contain the same keyword.


Why are the Impact Factors different from other sources?

We calculate the Impact Factors based on the citations registered in our database. We do not use the Impact Factor commercially released by Clarivate.

The concept of calculating impact factors is simple, but the two key elements are the number of articles and citations. Counting the number of citations is not the difficult part, but identifying the number of citable documents. Our basis for the citable documents is peer-reviewed articles determined by a native algorithm, which works well except for a few journals such as Science, which have different article types without specific features.

The corresponding IF Computation graph shows how we calculated the IF. If someone feels the IF is incorrect, that graph shows the root of the problem: whether we miscounted the citations or misidentified the peer-reviewed articles.

It is a misconception that the impact factor is a trademarked product.

Why are data subject to change?

We regularly update the database. If we find new citations or find some errors, the citations will be corrected, and thus, all the indices using those citations will be updated. It is not very common for old papers, but it can frequently occur for new ones.

What are the scientometric indices used throughout the website?

We will write detailed blogs describing the indices. However, you can find general information about them on Wikipedia or similar resources.

Is there any rate limit on using the service?

Just keep in mind that we have limited resources. When you overuse the system, someone else cannot use the system.

Why are some authors missed in the search results

We only list the authors for whom we have a profile. The purpose of displaying the name of authors is to assist researchers in examining their contributions.


Is there any rate limit on using the service?

Just keep in mind that we have limited resources. When you overuse the system, someone else cannot use the system.

Why is there no API?

We do not have the resources to handle high volumes of requests via API. Applying rate limit by authentication added to the complication and is not our priority.

Why are the datasets not available for downloading?

We regularly update the data, not simply adding new documents. We revise our algorithms to improve the data analysed throughout the system. Once again, we do not have sufficient resources to make all versions available. Furthermore, we do not wish to create havoc by distributing unversioned data, whose roots cannot be traced back.

We aim to release curated versions. However, instead of randomly releasing them, we plan to release them through peer-review publications, as the data are independently reviewed before release.