1 / 10

Review of AI from Chapter 3

Review of AI from Chapter 3. Journal May 13. What advantages and disadvantages do you see with using Expert Systems in real world applications like business, industry, medicine, etc.?. Review of Expert Systems. What is an expert system? What are some components of expert systems?

ailis
Download Presentation

Review of AI from Chapter 3

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Review of AI from Chapter 3

  2. Journal May 13 • What advantages and disadvantages do you see with using Expert Systems in real world applications like business, industry, medicine, etc.?

  3. Review of Expert Systems • What is an expert system? • What are some components of expert systems? • Knowledge base • If-then rules • Interface • How the user is interacting with the system • Process for matching, selecting, and executing the rules • This is called an inference engine

  4. Pattern Recognition • Identifying recurring patterns in input data with the goal of understanding or categorizing that input • Where have we seen Pattern Recognition before? • OCR • CAPTCHA • Remember CAPTCHA is an image, not text

  5. Optical Character Recognition • Locating and identifying printed characters embedded in an image, allowing the text to be stored as an editable document. OCR can be performed by wand readers, pen scanners, and OCR software.

  6. Steps for OCR • How do OCRs work? • First step is to scan the image of the page into the computer’s memory with a scanner, digital camera, or fax modem • Remember that the scanned image is nothing more than a series of bits (0s or 1s) in memory • Before the computer can process the text on the page, it needs to recognize the individual characters • OCR software scans the scanned image, locating and identifying each character, comparing them to a database of stored characters • When a match with close enough likeness is found, a text version of that character is then inputted into a new document

  7. Why is OCR not perfect? • Handwriting lacks consistency • Scan of paper may be of poor quality • Older documents may be faded or pages off color • Outdated database of characters • How do OCRs overcome this? • Segmentation of the page • Expert systems • Using knowledge base to recognize rules that distinguish character • Context experts • Feedback from humans-CAPTCHA

  8. Speech Recognition/Text-to-speech • These softwares also use AI in order to work properly • Speech recognition uses techniques similar to OCR • Segmentation of input sounds into individual words and phonemes • Expert rules for interpreting sounds • Context experts • Learning for human trainers • Remember that training speech recognition is becoming less common due to advances made by companies like Google • What does Google do?

  9. Image Analysis • One of the most difficult tasks to accomplish for computers is Image Analysis • Used for everything from: • Autofocusing cameras on humans • Piloting cruise missiles • Security programs (biometrics) • Space technology (Mars rover) • Robotics

  10. Complications with Image Analysis • Masses of irrelevant data • Objects that partially cover other objects • Indistinct edges • Changes in light and shadows • Changes in the scene as objects move • Face recognition

More Related