Docs

DataAgent (OnPremise)

Common Template: Case Introduction

Prediction of Automotive User Intent Tags

1. Scenario overview

Automotive marketing features a wide range of intent recommendation scenarios, such as common optional package intent recommendations, potential customer car purchase intent prediction, and replacement or maintenance intent prediction. Business personnel can leverage user lifecycle management to implement different marketing strategies for user groups at various lifecycle stages, for example:

For unaware users: automotive brand/model introduction, driving knowledge push
For users learning to drive: automotive brand/model introduction, car purchase needs understanding, driving knowledge, car purchase knowledge
For users planning to buy a car: automotive brand/model introduction, purchase timing, purchase budget, recommended model push, car purchase knowledge, automotive discount promotion
For users about to purchase a car: automotive brand/model introduction, purchase timing, purchase budget, recommended model push, car purchase knowledge, automotive discount promotion, lead collection
For users who have just purchased a car: automotive brand/model introduction, purchase timing, brand/model, automotive offer promotion, lead collection, car purchase knowledge
For existing car owners: automotive brand/model introduction, repurchase or replacement needs, purchase budget, intended car brand, replacement lead collection, automotive knowledge, offer promotion

This case focuses on scenarios involving users planning to purchase a car, users about to purchase a car, and existing car owners. It uses machine learning to infer users' car purchase intent, training on historical data from car purchase and replacement users, to predict purchase intent for users in these three lifecycle stages and implement appropriate marketing strategies.

2. Modeling process

2.1 Data description

Field name	Field type	Field meaning	Field description
uid	bigint	User ID
gender	string	Gender	Gender: male, female
age	string	Age group	Age group: 18-23, 24-30, 31-40, 41-50, 51-
area	string	Permanent residence	Permanent residence - area
lifestage	string	Life stage	Life stage: Seeking marriage, preparing for marriage, pregnancy, already have children
education	string	Education	Education: High school or below, associate degree, bachelor's degree, master's degree or above
job	string	Occupation	Occupation: International student, doctor, lawyer, teacher, blue-collar, white-collar, others
birthday	string	Birthday	Birthday
lifecycle	string	Lifecycle	Lifecycle: Fan, Potential customer, Car owner, Lost
own_store	string	Affiliated store	Affiliated store: Zhonghuan Guanghui, Pudong Yongda, Others
own_sales	string	Responsible sales representative	Responsible sales representative: Dawson, Wang Ning, Robert
wanted_store	string	Intended dealer:	Intended dealer: Pudong Yongda, Zhonghuan Guanghui
wanted_carmodel	string	Intended car model	Intended car model: RX5 Plus, ER6, RX3, i5, i6, iMAX8, Unknown
rank	string	Intention rating	Intention rating: High, Medium, Low, Completed transaction
lasttime_pudongyongda_testdrive	string	Last store test drive time	Last store test drive time: within 1 week, within 2 weeks, within 1 month, within 2 months, within 3 months, within half a year, within 1 year
newcar_tasting	string	New car appreciation event	New car appreciation event: Yes, No
storetimes_lastthreemonth	string	Dealer visit count in the past three months	Dealer visit count in the past three months: 0, 1, 2, 3, 4, >4
storenumbers_lastthreemonth	string	Number of dealers visited in the past three months	Number of dealers visited in the past three months: 0, 1, 2, 3, 4, >4
warranty_times	int	Store visits during warranty period	Store visits during warranty period: 0, 1, 2, 3, 4, 5
warranty_values	string	Total service value during warranty period	Total service value during warranty period: 0-3000, 3000-5000, 5000~10000, above 10000
pudongyongda_noentry_time	string	Duration not entering this store	Duration not entering this store (Pudong Yongda): 1 month, 2 months, 3 months, 4 months, 5 months, 6 months, 7 months, 8 months, 9 months, 10 months, 11 months, 12 months
forwardtimes_lastthreemonth	string	Total forwards in the past three months	Total forwards in the past three months: <3, 3-5, 6-10, >10
publishtimes_lastmonth	string	Post count in the past month	Post count in the past month: 0, 1, 2, 3, 4, >4
applets_opentimes_lastmonth	int	Mini program open count in the past month	Mini program open count in the past month
optionalpackpage_opentimes_lastmonth	int	Optional package page open count in the past month	Optional package page open count in the past month
optionalpackpage_time_lastmonth	int	Optional package page browsing duration in the past month	Optional package page browsing duration in the past month. Unit: minutes
last_eta	int	Current ETA	Current ETA. Unit: days
eta_modifytimes	string	ETA modification count	ETA modification count: 0, 1, 2, 3, 4, >4
pointuse_lastmonth	string	Whether points were used in the past month	Whether points were used in the past month: Yes, No
point_count	string	Points amount	Points amount: 0-500, 500-10000, 10000-30000, above 30000
active_lasttime	string	Last active time	Last active time
contact_preference	string	Contact preference	Contact preference
gift_preference	string	Premium category preference	Premium category preference
carowner	string	Car owner for this product	Car owner for this product
salescarmodel	string	Transaction car model for this product	Transaction car model for this product
salestime	string	Transaction date for this product	Transaction date for this product
carownerappvisit	string	Car owner app visit frequency	Car owner app visit frequency
othercarmodel	string	Other owned car models	Other owned car models
interest_car_price	string	Car price	Car price
interest_car_brand_country	string	Country of car brand	Country of car brand
interest_car_brand	string	Car brand	Car brand
interest_car_other_series_type	string	Car model	Car model
interest_car_newenergy_type	string	Type of new energy vehicle	Type of new energy vehicle
client_stage	string	Customer stage	Customer stage
channel	string	Information collection channel	Information collection channel
community_pv_lastweek	int	Page views in the community in the past week	Page views in the community in the past week
community_creation_lastweek	int	Community content generated in the past week	Community content generated in the past week
community_forward_lastweek	int	Community shares in the past week	Community shares in the past week
community_recruit_lastweek	int	Community new user acquisition in the past week	Community new user acquisition in the past week
car_optical_pack_eagerness	string	Optional package preference	Optional package preference
car_pick_up_eagerness	string	Car pick-up urgency	Car pick-up urgency
returned_customer	string	Returning customer	Returning customer
car_price_comparison	string	Price comparison across stores	Price comparison across stores
stored_many_points	string	Large point accumulation	Large point accumulation
club_freshman	string	Club newcomer	Club newcomer
recall_values	string	Churn recovery value	Churn recovery value

The data includes most of the key processed indicators, including user lifecycle, activity level, channel information, and personal attributes. By using such structured data and applying straightforward modeling, it is possible to predict intent ratings and thus enable precise marketing.

2.2 Data preparation

Data integration is achieved by directly connecting to Hive data. This case involves two datasets: historical manually labeled data and newly generated data. Classification training is performed on historical labeled data based on intention rating, and intention prediction is carried out on new data.

2.3 Problem modeling

Core metrics are selected for intention training. Since data processing has already been completed in advance, it can be understood that all metrics fundamentally influence the selection of intention. Therefore, there is no need for feature selection or feature processing. The target requirements can be met through simple logistic regression.

2.4 Model evaluation

Supports model evaluation functionality for training, assisting users in manual parameter tuning. Refer to the confusion matrix below, which shows the recall and precision of the final model performance. It displays the performance on the test set under different classification scenarios, and also supports the calculation of weighted recall, weighted precision, and accuracy.

2.5 Data production application

After obtaining a model with satisfactory performance, the predicted results can be exported to Hive/Clickhouse. Subsequent visualization queries and dashboard creation can be performed on the DataWind platform. It can also be integrated with CDP products for label insights, marketing applications, user segmentation, and more.

Last updated: 2026.06.02 14:59:34

DataAgent (OnPremise)

2.1 Data description #

2.2 Data preparation #

2.3 Problem modeling #

2.4 Model evaluation #

2.5 Data production application #

2.1 Data description

2.2 Data preparation

2.3 Problem modeling

2.4 Model evaluation

2.5 Data production application