Platform For AI:Use cases of Machine Learning Designer

Case

Description

Recommend products based on user and item features

This topic describes how to recommend products based on object features.

Use FM-Embedding for matching recall

This topic describes how to use the Factorization Machine (FM) and Embedding algorithms to generate feature vectors for users and items.

Create an FM recommendation model based on the Alink framework

This topic describes how to use the FM algorithm template provided by Machine Learning Designer to create an FM recommendation model.

Use collaborative filtering to recommend products

This topic describes how to use collaborative filtering to recommend products.

Use Bipartite GraphSAGE for matching recall

This topic describes how to use Bipartite Graph SAmple and aggreGatE (GraphSAGE) to obtain feature vectors of users and items for matching recall.

Use ALS to predict ratings of songs

This topic describes how to use Alternating Least Squares (ALS), a factorization algorithm, to predict the ratings users give to songs.

Case

Description

Implement public opinion risk control based on reviews from a food delivery platform

This topic describes how to implement public opinion risk control based on reviews from a food delivery platform.

Use graph algorithms to manage financial risks

This topic describes how to use graph algorithms to manage financial risks.

Predict credit scores by using a scorecard model

This topic describes how to use the financial components that are provided by PAI to create a scorecard model based on credit card billing statements.

Predict system anomalies by monitoring system metrics

This topic describes how to build a monitoring model for system metrics.

Monitor user churn

This topic describes how to use the user feature algorithm that is provided by PAI to create a model to monitor user churn.

Case

Description

Feature Engineering

This topic describes the feature engineering generated by customized recommendation algorithms.

DSSM vector recall

This topic describes how to implement vector recall for the Deep Structured Semantic Model (DSSM) generated by customized recommendation algorithms.

Rank

This topic describes how to implement rankings.

U2I2I recall based on etrec

This topic describes how to implement the u2I2I recall based on etrec.

Case

Description

Implement consistent click-through rate prediction in the batch and real-time modes

This topic describes how to implement click-through rate (CTR) prediction based on an Avazu dataset and deploy a workflow in which the Min Max Scaler Batch Predict, OneHot Encoder Predict, Vector Assembler, and FM Prediction components are batch run in sequence to Elastic Algorithm Service (EAS) as an online service.

Predict heart disease

This topic describes how to use data mining algorithms to build a heart disease prediction model based on the medical examination data of patients with heart disease.

Classify news based on text analysis

This topic describes how to use the text analysis components that PAI provides to build a news classification model.

Predict the repayment ability of agricultural loan applicants based on linear regression

This topic describes how to use linear regression to predict the repayment ability of agricultural loan applicants based on historical loan records.

Use the Binning component to implement the discretization of continuous features

This topic describes how to use the Binning component to discretize continuous features.

Collect statistics from population census data (old version)

This topic describes how to use population census data to build a statistical model. You can use the model to analyze the impact of academic degrees on income based on attributes such as the age, job type, and education level.

Predict the performance of students in examinations

This topic describes how to use logistic regression to generate a performance prediction model. You can use this model to predict the performance of students in an examination based on the family background of the students and their behavior at school. You can also obtain the key factors that affect the performance of students in examinations.

Automatically classify tags

This topic describes how to use the text analysis components that PAI provides to automatically classify product tags.

Build models to predict hazy weather

This topic describes how to build models to predict the hazy weather based on the analysis of weather data that was collected in Beijing over a one-year period. The models can be used to identify the pollutants that most often produce smog, which is measured based on PM 2.5 levels.

Build a model to predict the output power of a power plant

This topic describes how to use a preset template in Machine Learning Designer to build a model to predict the output power of a power plant.

Identify electricity theft and leakage

This topic describes how to use a pipeline template provided by Machine Learning Designer to build a model for identifying users who steal electricity or are involved in electricity leakage. This significantly reduces the inspection workload of electrical inspection staff and ensures normal and safe electricity usage.

Offline scheduling

This topic describes how to use the data mining components that PAI provides to perform offline scheduling for ad CTR prediction.

Use TensorFlow to develop an image classification model

This topic describes how to use TensorFlow to develop an image classification model in the PAI console.