In-depth mining and application of artificial intelligence machine deep learning and big data technology in football match prediction recommendation analysis

In-depth mining and application of artificial intelligence machine deep learning and big data technology in football match prediction recommendation analysis

As the largest sport in the world, football is also the most influential sport. The World Cup, the European Cup, including the five major leagues every year, thousands of people become their chasers, fascinated because of their love. In 2009, with the approval of the Ministry of Finance, competitive football games were successively launched across the country. Competitive sports quiz, as a game with a high winning rate on the market, has attracted a large number of fans. Competition lottery football has been developed in China for more than 10 years. During these 10 years, a large amount of football scores, football odds, football handicap and other event data have been accumulated. In the big data era of information explosion, data is value.
Keeping pace with the times, using the combination of big data and AI artificial intelligence technology, analyzing and predicting the probability of betting lottery football results with twice the result with half the effort, realizing intelligent one-key analysis, greatly improving the analysis efficiency and prediction accuracy.

Implementation idea:

1. In terms of data: the number of goals scored in major competitions in the past 15 years, handicap and other betting data.

2. Analysis technology: CURE, Chameleon and BIRCH, etc.

(1) Agglomeration method: consider each object as a cluster, and then continuously merge similar clusters until a satisfactory termination condition is reached;

(2) Splitting method: first attribute all the data to a cluster, and then continuously split the data sets with the smallest similarity to each other, so that the clusters are split into smaller clusters until a satisfactory termination condition is reached. According to the different measurement methods of inter-cluster distance, it can be divided into: minimum distance, maximum distance, average distance and average distance, etc.

(3) AI intelligence: artificial intelligence (Artificial Intelligence), machine learning (Machine Learning), deep learning (Deep Learning).
(4) Analytical tools: sourced from TanQiuZhe.com, those who are interested can search on Baidu, open source.

1. Russian Super League PFC Solge 3-0 Match time: 2021-08-16 23:59 -0.75 The main handicap is wrong

2. Denmark Super Dimensions 0-1 Match time: 2021-08-17 01:00 0.00 Handicap vs.

3. Malaga 0-0 Match time: 2021-08-17 02:00 -0.25 Host handicap vs.

4. Cartagena 1-3 Match Time: 2021-08-17 04:00 0.00 Handicap vs.

5. Portuguese Super League Bilaneses 1-2 Match time: 2021-08-17 04:15 -0.25 Handicap vs.

Algorithm performance:

• high intra-class similarity high similarity within the cluster

• low inter-class similarity low similarity between clusters

The measure of similarity is represented by the distance function d(i, j), and the distance function is generally different for different types of problems.