More and more people want to use AI to empower their companies, businesses, and products. But many people don't know about AI, so they don't know which problems are suitable for AI and which problems are not suitable.

This PDF will evaluate the feasibility of AI empowerment from four perspectives and provide a framework for thinking.

Non-technical people can understand this article without any obstacles.

Let's take a long picture to explain the main content that will be covered in the PDF. If you want to know more, please download the PDF.

PDF download address

What problem does this PDF solve?

Artificial intelligence is regarded as a "black technology" by many people. It can do some magical things, such as: playing Go is better than humans, playing games is better than humans, and the beauty effect is so good...

The most powerful companies on the planet regard AI as an important strategy for the entire company. Google, Microsoft, Facebook, Amazon, Tencent, Alibaba, Baidu, ByteDance...

Many big guys are also talking about artificial intelligence that will bring the next technological revolution. If you think about how the "Internet" has revolutionized, you can probably know how powerful this revolution is.

But the biggest question is: how do I use AI in the age of AI?

The above question is too big to answer. We need to focus on the problem: When I face a specific problem in the business, AI is also a solution idea. So, is this problem suitable for AI to solve?

So, this PDF solves a problem:

Is the specific problem I face suitable for AI? What aspects need to be assessed?

4 evaluation dimensions

The four evaluation dimensions are detailed in the PDF:

  1. data
  2. 特征
  3. Learn
  4. Black box


The biggest difference between artificial intelligence and traditional computer programs is that it is based on data.

This is also the underlying logic of artificial intelligence, so data is the most important resource in the field of artificial intelligence. So we need to evaluate the data dimension from three aspects:

  1. Is the data available?
  2. Is the data comprehensive?
  3. Is there much data?
3 elements of data evaluation

Download the PDF to view the full text, or to view the full text of the data via the link below:

'What you need to evaluate before using AI-data"


The basic principle of artificial intelligence is to find deep hidden features from a large amount of data, and then learn to complete specific tasks by judging the features.

Based on this principle, artificial intelligence should deal with more complex problems rather than simple problems. Judging the complexity of the problem can be judged from the following two dimensions:

  1. Number of features
  2. Certainty of features
Characteristic quadrant

Less features + weak certainty: suitable for manual solution

Less features + strong certainty: suitable for rule solving

More features + strong certainty: suitable for rule solving

Many features + weak certainty: "can be considered" AI solution

Download the PDF to view the full text, or to view the full text of the feature article via the link below:

'What you need to evaluate before using AI-features"


The previous two articles have explained that rule-based capability boundaries are small, and many practical problems cannot be solved by rule methods.Artificial intelligence can expand the boundaries of computer capabilities.

In addition to expanding the capability boundary, artificial intelligence also has a very important characteristic-Continuous learning and continuous improvement.

How to keep the machine learning?

In order for the machine to achieve continuous learning, we need to implement 2 conditions:

  1. Continuously get feedback data to let the machine know where it is good and where it is bad
  2. If feedback data is added to the closed loop, can the machine learn continuously and improve its capabilities?

Download the PDF to view the full text, or to view the full text of the study via the link below:

'What you need to evaluate before using AI-learning articles"

Black box

Most of our computer science in the past was rule-based, much like a car, and we knew exactly how the car was assembled, so when we found that the screws were loose, we tightened them, and we replaced them with any parts that were old. Can be done right.

Deep learning is completely different. When we find a problem, we can't do it right. We can only optimize it globally.(Such as filling more data).

Case position in 2 evaluation quadrants

Therefore, there are 3 principles in the evaluation:

  1. The more the solution needs to explain the reason behind it, the less suitable it is to use deep learning
  2. The lower the tolerance for errors, the less suitable it is to use deep learning
  3. The above 2 clauses are not absolute judgment standards, but also need to look at commercial value and cost performance. Autonomous driving and medical treatment are counter examples.

Download the PDF to view the full text, or to view the full text of the black box via the link below:

'What you need to evaluate before using AI-black box"

All the above contents have been compiled into a 41-page PDF "Evaluated before the introduction of AI". Click the button below to download.

Download PDF "Need to be evaluated before introducing AI"

If the image above shows a problem, please click here to download:Download link