Business Solutions:
0. Data source Loading
1. Feature Engineering: Character-to-value/two-valued/multivalued-type features are converted into numerical representations that can be processed by the algorithm, and feature abstractions are realized. Features are two-value types, such as sex, which has male and fem two, which abstracts sex into 0 and 1. If the value of a feature is a multivalued type, such as status, it is abstracted from 0 to 1, and then to the second degree of severity.
2. Data preprocessing: Numerical double/normalized to 0 and 1 through the type conversion component, the data type is converted into double type (machine learning algorithm generally supports double type data), then the data is processed by "normalized component", Normalized all values to between 0 and 1
3. Training and Evaluation: Split component ", the data in the assembly randomly split 70% of the data according to 7:3 to train the model, 30% of the data used to predict.
4. Model evaluation: Because this experiment is a two classification scenario, the results of predictions and real values have been obtained through the "predictive component", but we need to be more intuitive to verify the accuracy of the experiment, so we selected the "Two classification evaluation component" to evaluate the results.
News text Analysis
1. Data preprocessing and word segmentation, adding serial numbers, participle
2. Keyword extract Word frequency statistics component
3. Article classification. "Ternary group to KV component" is a common algorithm for text vectorization, the principle is to convert text data into K:V format display,
Assault-data structure and algorithm crash
Lesson One: General application of stacks, queues, and lists from the basic data structure
Lesson two: Basic data structure ———— stacks, queues, list of kinky tricks
Lesson Three: ———— heap of advanced data structures, conventional application of binary tree
Lesson Four: The ———— of the advanced data structure and the deformation of the tree
Lesson Five: ———— of advanced data structures hash table, search tree
Lesson Six: Ordering ———— of compulsory algorithms
Lesson Seventh: ———— Division of the compulsory algorithm
Eighth lesson: ———— Greed of the compulsory algorithm
Nineth Lesson: ———— Search of compulsory algorithms
Lesson Tenth: ———— topological ordering of graph algorithms, minimum spanning tree (Kruskal and Prim)
11th Lesson: Graph Algorithm ———— single-source Shortest path (Dijstra, Bellmanford, SPFA) and its variable use
Business Solutions/-data structures and algorithms crash