Semantic scholar что это
Институт Пола Аллена запустил поисковик Semantic Scholar специально для учёных
Институт искусственного интеллекта Пола Аллена (Allen Institute for Artificial Intelligence, AI2), созданный одним из основателей Microsoft Полом Алленом, объявил о запуске поисковой машины Semantic Scholar, которая ориентирована на «умный поиск» в научных статьях. Пока что проект имеет статус беты и база данных документов содержит тексты только из области компьютерных наук. Основным отличием от конкурента Google Scholar заявлена возможность «семантического понимания», под которой имеется ввиду, что система может извлекать из статьи характерные элементы — ссылки, цитаты, ключевые фразы — и помогать пользователю ориентироваться в них.
Поисковая машина имеет типичный интерфейс, включая умные подсказки с именами авторов, доступными пока что только на английском языке. Поисковые результаты выглядят как список заголовков статей аннотаций к ним; в левой части доступны инструменты для анализа результатов. Помимо достаточно стандартных для такого рода поиска данных, таких как даты публикации или издательство, интересным выглядит блок «Key Phrase», который группирует «ключевые фразы», встречающиеся для предмета поиска в найденных статьях. При клике на ключевую фразу система показывает документы, где она, по мнению системы, играет ключевую роль. Найденные статьи свободно доступны в формате pdf.
Если попробовать перейти по какой-либо найденной ссылке, то система показывает список источников, использованных в документе. Они по умолчанию упорядочены в соответствии со степенью их «влияния» на текст. За определение степени «влияния» отвечает специальный алгоритм, работу которого в своей статье описал CEO AI2 Орен Эциони (Oren Etzioni). Соседняя кнопка показывает краткие выдержки из текста в виде всплывающих подсказок.
Пока что базы данных документов даже по компьютерным наукам у Semantic Scholar и Google Scholar несравнимы по количеству текстов: поиск по слову «java» в системе Google даёт 2 060 000 результатов, тогда как у новой системы результаты гораздо скромнее — 89 928 документов.
Tutorial Videos
Learn how to use our AI-powered academic search engine
Quick Links
101: Citations Overview
With billions of citations, Semantic Scholar provides a scientific literature graph that allows scholars to navigate and discover the most relevant research across all fields of study.
Our novel citation features allow you to discover highly influential works and easily search a paper’s citations. This video covers:
View Transcript
Hide Transcript
«Semantic Scholar is an AI-powered research tool for scientific literature. With billions of citations, Semantic Scholar provides a citation graph that allows scholars to navigate and discover the most relevant research across all fields of study. We offer many ways to explore our citation graph, including our novel method of classifying citations by type and influence, as well as searching citations.
Our system indexes billions of citations, using AI models to classify the intent and predict the influence of each. We’ve identified three classification types; cites background, cites methods, and cites results. You can find citation type by scrolling to the citations listed for any paper and using the citation type filter to identify papers that cites background, cites methods, or cites results.
While many citations are incidental, Semantic Scholar classifies some citations as being “Highly Influential,” allowing scholars to quickly determine which publications to read in depth. Find highly influenced citations sorting by Most Influenced Papers or scanning for the Highly Influenced paper badge under each paper title.
Searching citations of a paper helps researchers quickly navigate through a large list of citations and narrow results down to the most relevant. To get the most out of this feature, search for a keyword in the paper title, filter by date range, citation or publication type, author, or, filter by a specific conference or field of study. You can also sort by relevance, most influenced papers, citation count, or recency.
Explore citations to discover new research on semanticscholar.org.»
101: Search
Navigate through millions of scientific papers in all fields of study by searching our corpus.
In this video learn search best practices like:
View Transcript
Hide Transcript
«Semantic Scholar is an AI-powered search and discovery tool used by millions of scholars globally each month, and supports hundreds of thousands of scholarly searches each week.
Our mission is to empower researchers to fight information overload. Searching on our platform enables scholars to navigate through hundreds of millions of papers in all scientific domains.
So, how does it work?
Underlying all of Semantic Scholar is a deep semantic analytic engine that helps researchers understand the meaning of a paper. Natural language processing and machine learning models rank the results by relevance, all to help you quickly find the most up-to-date research in your field.
To get started, visit semanticscholar.org and enter a search for a paper, author, topic, or keyword. You can further refine your search results using our many filter options.
Select field of study to find literature from a particular domain.
If you’re looking for recent research, filter your results down to the past two years.
You can also filter by publication type, authors, or journals and conferences.
Clicking on a paper title will take you to our paper pages. To learn more about Semantic Scholar paper pages, watch the tutorial. Find new research at semanticscholar.org.»
101: Paper Pages
Semantic Scholar’s corpus includes nearly 200 million academic papers across all domains, including biomedicine, computer science, business, history, and economics.
Our system extracts key information from papers and presents it on the Paper Page for review. Learn how to:
View Transcript
Hide Transcript
«Semantic Scholar is a free, AI-powered research tool for scientific literature used by more than 8 million scholars globally.
With nearly 200 million papers across all scientific domains, Semantic Scholar provides professional scholars like researchers, engineers, clinicians, and scientists with free, AI-powered tools to quickly find the research that is most relevant to their work.
To get started, visit semanticscholar.org and search for a paper, author, topic, or keyword. You can filter by field, date range, conference, and more to refine the search results as well as sort by relevance or recency.
On a paper page, you can browse videos, presentations, code libraries, and clinical trials. Get a glimpse at a paper’s figures, tables, and topics, skim news or blog articles discussing a paper, and navigate citations and references to find connections to other papers.
About Semantic Scholar
Helping scholars discover new insights
We are a team of researchers and engineers at the Allen Institute for AI building a better way to search and discover scientific knowledge.
Scholars Building For Scholars
Staying up-to-date with scientific literature is an increasingly pressing challenge for scholars.
Our mission is to accelerate scientific breakthroughs using AI. Our Research team studies information overload and develops AI tools to overcome it. And our Product and Engineering teams deliver these unique AI-driven features at scale to millions of scholars each month.
Semantic Scholar is and always will be open and free for all to use. In addition to the search and discovery tools available at semanticscholar.org, we provide the Semantic Scholar API and Open Research Corpus as free services to the research community.
AI For The Common Good
At AI2, our mission is to contribute to humanity through high-impact AI research and engineering.
Semantic Scholar was launched in 2015 as a groundbreaking project at the Allen Institute for AI, a non-profit research institute founded in 2014 with the mission of conducting high-impact AI research and engineering in service of the common good. AI2 is the creation of Paul Allen, Microsoft co-founder, and is led by Dr. Oren Etzioni, a leading AI researcher.
Semantic Scholar Advisory Board
The Semantic Scholar Advisory Board provides ongoing guidance as we seek to help scholars around the world overcome information overload.
Ilias Nikolakopoulos, MD
Internal Medicine Resident
Yale School of Medicine
David E. Cummings, MD
Professor of Medicine
University of Washington
Sravanthi Parasa
Gastroenterologist, Clinical Researcher
Swedish Medical Center
Elena L. Glassman
Assistant Professor of Computer Science, Harvard University, and Stanley A. Marks and William H. Marks Assistant Professor, Radcliffe Institute
Bryan Pardo
Professor of Computer Science
Northwestern University
Are you interested in our mission? Join the team!
Partnering with 50+ Publishers and Scholarly Societies
Reach millions of scholars by integrating your academic content into Semantic Scholar’s knowledge graph
Help improve the future of scholarly research
Join our Beta Program to provide feedback and help scholars around the globe
Semantic Scholar API
Providing a reliable source of scholarly data for developers
Build projects that accelerate scientific progress with the Semantic Scholar API
Driven by State-of-the-Art AI Research
We provide the RESTful Semantic Scholar API as a service to the global research community. The API is a reliable on-demand source of data about authors, papers, citations, venues, and more that allows you to link directly to the corresponding page on semanticscholar.org for more information.
Driven by state-of-the-art AI models developed by our Research team, the Semantic Scholar API powers applications that enable scientific discovery. Currently the API supports Paper and Author Lookup, Conflict of Interest detection, Conference Reviewer Match, SPECTER embeddings, and SUPP.AI annotations. We are actively developing new features based on user demand.
Paper and author lookup are available from our Literature Graph service, whose reference documentation is available here. Documentation for our Conference Peer Review service is available here.
Semantic Scholar providing infrastructure for the research ecosystem is exciting as it enables partners like Litmaps to focus on creating great experiences for end users. We found the Semantic Scholar API and bulk data, clear and straightforward to use. The documentation, examples, and easy access to the raw data allowed us to move quickly on data integration, and get back to solving problems for our users.
Axton PittCo-founder and CTO, Litmaps
Since day 1 Connected Papers has been built on top of the Semantic Scholar Open Corpus and APIs, and heavily relies on them to provide paper discovery and search capabilities to our users. The APIs are well maintained, frequently updated, fast, and easy to use. The team is extremely responsive and has supported our growth every step of the way from open beta to a million users. This massive resource given to the public for free is a great driver for innovation in science and we are proud to be a part of the next layer of tools being built on top of it.
We are a startup that tracks the evolution and progress of AI research. Thanks to the Semantic Scholar API pulling in the latest references to papers from top AI conferences has been a really simple task. Otherwise, it would have taken much more time and effort.
Eduardo Antonio Espinosa Grimaldo Founder, Stateoftheart AI
ИИ Semantic Scholar сократит длинную научную статью до одной фразы
Институт искусственного интеллекта Аллена в Сиэтле на минувшей неделе представил систему Semantic Scholar, которая предлагает экстремальное сжатие пространных научно-технических статей, чтобы сократить время на их изучение.
Это исследовательский инструмент на базе ИИ, который может пригодиться в научной среде. Благодаря функции реферирования он может просматривать огромное количество научной литературы, сводя обработанный материал буквально к одному предложению.
Услугами Semantic Scholar за месяц воспользовались 7 млн. пользователей, в его базе находятся 10 млн. статей по информатике. По словам одного из разработчиков системы Дэна Велда, существующая база данных будет регулярно пополняться – в том числе статьями и по другим дисциплинам.
Это далеко не первая программа обработки естественного языка для обобщения документов. Обычно для решения этой задачи используется один из двух подходов – либо экстрактивный, делающий выбор в пользу репрезентативного текста и его дословного использования в резюме, либо абстрактный, когда используются алгоритмы генерации естественного языка при создании резюме с оригинальной формулировкой.
В отличие от них у Semantic Scholar чрезвычайно высокая степень сжатия. К примеру, резюме статьи из 5000 слов составит всего 21 слово, то есть 1:238. У ближайшего конкурента Semantic Scholar этот показатель составляет 1:36.