Machine learning can be said to be hot. As long as you apply machine learning, you can effectively enrich your data and knowledge and facilitate the automation of valuable tasks, including perception, classification, and numerical prediction. And its "brothers" - machines found, can be used to discover new knowledge that illuminates and guides humans. Let's explore the best application scenarios for machine learning or machine discovery, and why they are important to business.
Many years ago I was a machine discovery researcher, published academic papers in Machine Learning magazine, participated in machine learning related conferences and made reports, because machine learning and machine discovery were similar to human activities. As an (experienced) entrepreneur, I am often asked if the learning method is very important for automating certain tasks, which is why I am writing this article. Let us first review some basic concepts.
An important idea in the field of artificial intelligence is that intellectual work can be seen as a heuristic search in the "problem space" that can help find solutions to problems.
Let us imagine a common mission scenario in TV: the squadron agent arrived at the crime scene and the body lay on the floor. A bad detective picks up the phone book and calls from the first page to conduct an interrogation. A very bad detective would even think that it was the invasion of space or escape, and asked NASA and the local zoo to pursue these clues. Their behavior is to use the wrong heuristics.
A good detective will be good at using the right heuristics to start with existing problems, such as: What is the cause of death? Who is the last person the victim saw? Is there an enemy? Is there a secret romance? Debt? Good detectives will also start with the above answers to more effectively search for suspects on a large scale. Great detectives may even come up with more inspiring ideas.
The key point of "machine discovery" is that discovery is like another intellectual task, so the artificial intelligence key heuristic discovery search method applied in the problem space can also be applied to machine discovery tasks.
On the other hand, the key to "machine learning" is to give enough data and related results, as well as concepts (such as which data features are related to the predictions), and then the software can be trained to achieve this association. Classic examples include using historical data to learn how to classify loan applications based on credit risk, or to predict customer churn.
What is the best application for machine learning or machine discovery?
With these key points, let's consider which design (discovery or learning) is better in a specific application. For example: Introducing traffic for large parties or events. A good party organizer needs to understand the common interests among the guests and try to introduce them to each other and explain their commonalities to promote communication between them. This is a daunting task, so the organizers are very busy. Can this be automated with a list of participants?
Artificial intelligence or discovery methods deal with things like this: research or find out what can lead to good mutual introduction. What determines the quality of the (referral)? Is this an innovative introduction to the core purpose? Which data sources can enhance this automatic referral (such as LinkedIn or other self-introduction)?
Then, you can generate some automatic introductions, for example: you three graduated from the same university almost at the same time; or you have served the African Peace Organization; even two of you are the only ones who know machine learning here.
Bad ways to inspire may lead to: You are divorced more than four times (å°´å°¬); or you are all from the Midwest (with a focus on blurring); or your birthday is in winter (not relevant).
We have discussed the key points of machine learning and machine discovery, and how to implement specific applications. So let's summarize: What is the best application for machine learning or machine discovery?
Machine discovery requires the logic of the research task, requires knowledge, includes priority paths within the scope, and makes it conform to the actual algorithm design. This facilitates the innovation of the space being searched and the heuristics used. But the biggest innovation may come from the novel, creative output that is based on specific inputs, because automation can explore the possibilities of much larger space than humans actually consider.
Let's take a look at three examples of machine discovery engines, each of which uses program-programmed heuristics to explore and report as much human-readable knowledge as possible.
In the 1990s, commercial search engines searched for a lot of information files, using heuristic techniques (such as page sorting, prioritization based on the content of each document or the query term of the title) to give a list of citations, and each excerpt was dynamic. Customized as a function of the query word.
The commercialized categorization engine around 2000 puts hundreds of search results in grouped form into topic folders, using heuristic techniques (such as the language features of the extracted topics, how many search results each topic covers, The topic is divided into the effects of non-overlapping groups, etc.) to describe the topics that appear in the returned search results.
In 2015, the commercialization benchmarking engine found its anomalous performance in large similar groups, using heuristic techniques (such as combining simple and reasonable attributes, and dealing with anomalous types of well-formed sentences) to output the ability to convey benchmarking insights about the target entity. English passage.
Perhaps the way the machine discovers is that the task output is not just a classification or a numerical prediction. People have written a lot of books or articles about this kind of task to teach new people. There is also no rich data on input/correct output groups, so it is often convincing why others input data and task metacognition knowledge meet specific outputs. Task metacognitive knowledge is isolated, so general knowledge is not required when performing tasks.
What does this mean for technology business? Machine learning enables automated tasks to be semi-automated to reduce expenses. Machine learning can be applied to many data-rich tasks. Machine discovery emphasizes specific tasks that require specific knowledge and training. Machine discovery tends to be hand-crafted, more elaborate and rare.
You need a lot of internal or vendor artificial intelligence expertise. There are fewer suppliers, and they are more focused on specific knowledge tasks with far-reaching impacts to ensure that the business is economically viable. Suppliers do not call themselves a machine discovery company. Unlike machine learning, because machines find fewer companies, they are more likely to differentiate in the market.
Machine learning and machine discovery are brothers, but they are separated when they are mature.
Heating Pad For Feet,Heating Pad For Foot,Foot Heating Pad,Cordless Foot Warmer
Ningbo Sinco Industrial & Trading Co., Ltd. , https://www.newsinco.com