Dilemma: Easier said than done

Not long ago, Lin Hui was thinking about labeled data and dark data.

Although this kind of thing has great direct value and incidental value.

But it is also difficult for Lin Hui to move it.

From a purely technical perspective, there are a lot of troubles.

Lin Hui is aware of many commonly used data mining methods.

But it seems that it is very difficult to reproduce now.

Take the big data mining method based on AI and cloud computing that was commonly used in mining data in previous lives.

Judging from the name, this lousy data mining method uses three of the most popular concepts in the computer/Internet field in the past.

——Artificial intelligence, big data, cloud computing.

Indeed, this method of big data mining based on AI and cloud computing is closely related to the above three.

Follow this approach for data mining.

First, it must be applied to a cloud service center that has communication connections with multiple online service terminals.

When using this method for data mining, it is also necessary to obtain the mining evaluation index information of the big data mining business corresponding to the big data service control that can currently be executed for the big data decision-making information of the online cloud computing project.

When specifically conducting data mining, the artificial intelligence model must be configured in advance.

As for why you need to configure an artificial intelligence model?

Because only if the artificial intelligence model is configured, the index classification of the mining evaluation index information can be achieved based on the pre-configured artificial intelligence model.

In this way, it is easier to obtain the indicator classification results.

Getting the indicator classification results is not enough.

On this basis, the indicator classification results should be further divided into multiple indicator classification sets.

Then the corresponding indicator classification mining features are extracted from the multiple indicator classification sets divided by the indicator classification.

Only in this way can efficient and accurate mining be achieved.

If you want to achieve efficient big data operation efficiency in the future.

In the process of data mining, indicator classification mining features are used in addition to providing certain quantitative data.

It is also used to represent the clustering theme features corresponding to the clustering theme cluster corresponding to the indicator classification set.

This in turn requires determining the mining service model between each indicator classification set based on the extracted indicator classification mining features.

And on this basis, the mining service model between each indicator classification set is determined.

That’s not all, we still need to build the corresponding mining service topology map later.

According to the constructed mining service topology map.

Only in this way can the big data mining process corresponding to each indicator classification set be determined.

After determining the big data mining process corresponding to each indicator classification set.

According to the big data mining process corresponding to each indicator classification set and the subject entity relationship with subject category identification between the multiple indicator classification sets.

Then you can execute the big data mining process corresponding to each indicator classification set in the indicator classification results.

Lin Hui’s steps above are already quite basic.

In fact, the steps to construct the corresponding mining service topology map based on determining the mining service model between each indicator classification set are far more simple than what can be described in a few sentences.

It actually involves:

How to determine the mining service model between various indicator classification sets?

How to divide each target indicator classification set covered by the same type of mining service model into a mining service distribution map?

How to reduce the distribution range of the mining service distribution map whose distribution heat map matches the preset thermal characteristics according to the distribution heat map in each mining service distribution map, and expand the distribution range of the mining service distribution map whose distribution heat map is less than the preset quantity threshold.

Get the adjusted distribution map of each mining service?

These problems are not easy to solve overnight.
Chapter completed!