AITopics
AITopics is the Internet's largest collection of information about the research, the people, and the applications of Artificial Intelligence. Our mission is to educate and inspire through a wide variety of curated and organized resources gathered from news, conferences and publications.
AITopics 3.0 is brought to you by The Association for the Advancement of Artificial Intelligence (AAAI) and is powered by AI technology from i2k Connect. The new engine combines machine learning with subject matter expert knowledge to automatically tag documents with unique, accurate and consistent metadata. Building on the familiar online shopping experience, you can discover and analyze just the information you need.
Click on any page for help or display the complete user manual.
Quick Links
When you first arrive, you will see the latest information.
Search for a Specific Document / Browse for Documents on a Specific Topic
-
Keywords. Enter words or quoted phrases into the search box in the top navigation bar and press Enter. Review the results presented below and refine as needed using additional filters in the left-hand-side panel.
-
Filters. To open any of these filters, click the icon to display category names under each filter. Click on one of the names to restrict the search further. There may be sub-categories for each category, found with the arrow to the left of the category. You can drill down into smaller sub-categories by clicking each arrow. The numbers in parentheses indicate the number of documents associated with each filter selection.
-
Current Filters. You will see the Current Filters box on the top left display each filter you have applied. If you no longer need a particular filter, click the beside it in the Current Filters box to remove it. If the Current Filter becomes complex, the i2k Connect system will automatically convert it to a Focal Point.
-
Source. The i2k Connect Platform scans content from hundreds of sources (RSS feeds, Web sites, Blogs, Twitter feeds), as well as the i2k Connect reference library file system. Select from the Source menu to focus on one. Alternatively, select
Theme to focus on a set of related sources.
-
Date. Choose a date in the date box or click the
icon to enter a date range. You may also click the icon to see a histogram of documents by publication year. Slide the circles along the
axis to tighten the focus. Note that when you restrict a search by date, the dates of the documents included in the search filter are displayed in blue, while any documents excluded by the filter are shown in grey.
-
Theme. Each source can have one or more themes. Different sources can also have the same theme. For instance, your system may have many sources – Businessweek, The Economist, Financial Times, Yahoo Finance – that all have a Financial News theme. Searching with the Financial News theme will search all Financial News sources at the same time, without needing to manually pick every source individually.
-
Author. Click to focus on documents that are written by a particular person.
-
Concept Tag. The i2k Connect Platform identifies important
words or phrases found in a document. Select from the Concept Tag menu to focus on words or phrases you believe are important in the document(s) you are looking for. You can also click on the icon to explore the concepts identified in a set of documents.
-
View. The i2k Connect Platform classifies documents into topics arranged in hierarchical taxonomies.
Select from a menu with a icon to focus on a topic.
Once you have selected a topic, you have the option to zoom in on Overviews, Instructional Materials, past weekly AI-Alerts, or Classics about the topic.
Click News to return to the default display.
Available Views
The i2k Connect Platform classifies documents into topics arranged in hierarchical taxonomies called Views.
-
Country. Regions and countries of the world; including level 1 & 2 subdivisions (e.g., state, province, county).
-
Genre. Forms of business and technical writing; e.g., instructional material, overview, press release.
-
Industry. Industrial sectors (e.g., Oil & Gas, Freight & Logistics Services), drawn from a simplified version of the Global Industry Classification Standard (GICS).
-
Technology. Information technologies, with an emphasis on Artificial Intelligence.
-
Conference. Click to focus on documents from a particular conference.
-
Journal. Click to focus on documents from a particular journal.
-
File Type. Web Page, PDF, Microsoft Word, ... This helps you narrow the search to exactly the type of file you need.
When you use this site on a mobile device, tap the icon at the top of the screen to show or hide the search box, and tap the icon to show or hide the left-hand-side panel.
Top Navigation Bar
-
Home. If you need to start over, press the Home button in the top navigation bar to begin a new search.
-
About. Background information on AITopics, including articles about the system.
-
A Brief History of AI. From ancient history through the 2000's.
-
AI-Alerts. Weekly digest of artificial intelligence news from around the world – automatically selected and enriched by AI technology. Includes a subscription link.
-
AI Magazine. Articles from the AAAI flagship publication – the "journal of record for the AI community."
-
AAAI Conferences. Articles from AAAI conferences.
-
Books. Sponsored AI-related books from publishers.
-
Classics. Classic books and articles about AI.
-
Views. documents are classified through the lenses of several hierarchical Views (taxonomies). Use the selector to decide which to display with your search results and in the filter selection menu. The selected Views will stay with your profile until you change them.
The available Views are described in the Filters section.
Search Box Syntax
Search for documents that contain all of the words you enter, or use this advanced search syntax.
| Exact phrase | | "This That" |
| Boolean operators | | This AND That ( or This That ) |
| | This OR That |
| | This NOT That |
| Fields | | title:"Coalbed Methane"
pagetext:environmental (document contents)
filepath:JPT (string contained in the URL)
concept-tags:"drilling technology"
store:"health news" (e.g., name of news feed)
summary:"social responsibility"
views:"Technology|Information Technology|Artificial Intelligence"
(a topic in one of the selected taxonomy lenses)
|
| Wildcards | | title:Coal* ( Find documents with words in the title that start with "Coal" ) |
| Proximity | | "crude train"~3 ( Find documents where 'train' and 'crude' occur within 3 words of each other; e.g., "train carrying crude". ) |
Research Dashboard
Research Dashboard. Click on one of the four panels – Timeline, Concept Tag Cloud, Document Map, or Collaborating Authors. Turn the dashboard on or off by clicking on the icon and clicking on the "Dashboard" checkbox.
- Timeline. Shows the number of documents found for each year in blue. Narrow the search range by either typing dates in the white boxes or by sliding the round white buttons along the bottom edge. Press Go to enter. Hover over any column in the histogram to see the count for that range. Note: When a search is active, those items included in the search are displayed in blue, and those excluded are in gray.
- Concept Tag Cloud. Important words or phrases found in all documents known to the platform, displayed as a cloud. A search will change the cloud to include only tags in those documents. The size of the words indicates the number of times the tag appears in the corpus. Larger fonts indicate more popular tags. Each tag is a link that can be clicked on to narrow the search to that topic. The size of the words can be adjusted using the toggle switch and sliding the blue circles up and down along the side.
- Documents Map. Search documents that are associated with certain areas by clicking on the pencil icon and drawing a rectangle on the map. Select layers to turn on or off with the layers icon.
- Collaborating Authors. Shows author names as dots with lines between collaborators. Larger dots indicate more documents associated with that author. Color indicates publishing date.
Results Display
Search results are shown in the right column on large displays and cover the screen on small displays.
The newest documents are displayed first by default. You may also select Sort By Relevance or Sort By Title.
Actions. Click for options to explore the items in the current search results.
-
Export to Excel. Mouse over to select the number of items to export.
-
Other Dashboards.
- Default: No special dashboard.
-
Sunburst Views: Illustrates the proportion of classifications into the different topics of one View.
-
Show as Treemap. Click for a graphical display of the distribution of items. Click on any box to see more detail.
-
Quick Topic Analysis. Click to create charts that show the distribution of classifications of items.
Click the icon to show the current search results on the map.
Click the icon to see detailed information about any document.
Information displayed for each document includes:
-
Title. For a Web page, the title is identified from the Web page metadata. For files like PDFs and Microsoft Office documents, the title is identified from the content of the file.
Clicking on the title will open the document in a new browser tab.
On a small-display device, such as a phone, click next to the title to see the additional metadata.
-
Author. The owner of the file. Click to see only documents with that author.
-
Theme and Age. Where the document was published and how long ago.
-
Summary. An auto-generated summary is displayed for each document. If you have entered words into the search box, a snippet of text is displayed with the words highlighted in context.
-
Concept Tags. Important words or phrases found in the document. Click one to focus on documents that include the same concept.
-
or or or Source. News, Twitter, file or web page. Click to focus on that source.
-
Topics. How the document is seen through the "lens" of each selected hierarchical View (taxonomy). A confidence factor is shown for each topic. There may be several topics or none. Click one to see all documents identified for that topic.
-
Add feedback. Tell us about your overall impression of the new experience or about incorrect or missed topics, concept tags and other metadata for this document.
-
More like this. See similar documents. The popup menu lets you choose text similarity, topic classification similarity, or concept tag similarity.