MIT research Team develops "human-like" pattern Recognition model

The study is based on the thought that humans tend to think and make decisions based on their experiences and the examples they see. For example, a child may know from several words spoken by his parents that they are talking about summer camp because they have been there last year and they know that words such as "month," "Lake" and "counselors" will only be used together in this situation.

However, if we have limited experience or perhaps no experience in a particular field, a little help may be necessary-this is where Bayesian case model works. Given a set of data, such as recipes, the model is sorted based on their most prominent constituents and on representative instances or prototypes in a given recipe set, which is also selected by the computer.

For example, although I don't know about the same ingredient in beer, paprika, and ketchup, when I find out that the model is considered a typical chili recipe, I can conclude that the recipe contains chili peppers. In fact, MIT researchers been Kim, Cynthia Rudin and Julie Shah found that not only did their models perform more accurately than previous techniques, but human testers used Bayesian case The output of model technology can significantly improve the rate of sorting recipes than before.

Julie Shah (left) and Been Kim

This technique should use more complex types of data in certain areas.

Even if it is not the pattern itself, this type of work can be very useful as the dataset exceeds the analytical power of the population. Unsupervised machine learning and artificial intelligence models, for example, from software AYASDI and Google's famous cat face recognition depth learning system have been able to use a lot of data and identify similar things, but any tool can only be accurate and simple to help humans identify what it finds useful. Complete paper click here.

Original link: Researchers build pattern-recognition model which acts like a human (Zebian/Wei)

The 2014 China Large Data Technology Conference (Marvell conference 2014,BDTC 2014) will be held at Crowne Plaza Hotel, New Yunnan, December 12, 2014 14th. Heritage since 2008, after seven precipitation, "China's large Data technology conference" is currently the most influential, the largest large-scale data field technology event. At this session, you will not only be able to learn about Apache Hadoop submitter uma maheswara Rao G (a member of the project Management Committee), Yi Liu, and members of the Apache Hadoop and Tez Project Management Committee Bikas Saha and other shares of the general large data open source project of the latest achievements and development trends, but also from Tencent, Ali, Cloudera, LinkedIn, NetEase and other institutions of the dozens of dry goods to share. There are a few discount tickets for the current ticket purchase.

Free Subscribe to the "CSDN large data" micro-letter public number, real-time understanding of the latest big data progress!

CSDN large data, focus on large data information, technology and experience sharing and discussion, to provide Hadoop, Spark, Impala, Storm, HBase, MongoDB, SOLR, machine learning, intelligent algorithms and other related large data views, large data technology, large data platform, large data practice , large data industry information and other services.