Paper: arXiv 2403.14483

Authors: Shaojie Li, Xinqi Dong, Danqing Ma, Bo Dang, Hengyi Zang, Yulu Gong

Abstract

Mobile Internet user credit assessment is an important way for communication operators to establish decisions and formulate measures, and it is also a guarantee for operators to obtain expected benefits. However, credit evaluation methods have long been monopolized by financial industries such as banks and credit. As supporters and providers of platform network technology and network resources, communication operators are also builders and maintainers of communication networks. Internet data improves the user’s credit evaluation strategy. This paper uses the massive data provided by communication operators to carry out research on the operator’s user credit evaluation model based on the fusion LightGBM algorithm. First, for the massive data related to user evaluation provided by operators, key features are extracted by data preprocessing and feature engineering methods, and a multi-dimensional feature set with statistical significance is constructed; then, linear regression, decision tree, LightGBM, and other machine learning algorithms build multiple basic models to find the best basic model; finally, integrates Averaging, Voting, Blending, Stacking and other integrated algorithms to refine multiple fusion models, and finally establish the most suitable fusion model for operator user evaluation.

Complexity vs Empirical Score

  • Math Complexity: 4.5/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The paper focuses on practical machine learning implementation (LightGBM, ensemble methods) with moderate mathematical depth, but demonstrates strong empirical rigor through its use of massive real-world operator data, detailed data preprocessing, feature engineering, and comparison of multiple modeling techniques.

Research Flowchart

  flowchart TD
  A["Research Goal: <br>Mobile Internet User Credit Assessment <br>for Communication Operators"] --> B["Data & Inputs: <br>Operator-Sourced Big Data"]
  B --> C["Key Methodology: <br>Preprocessing & Feature Engineering"]
  C --> D{"Computational Process"}
  D --> E["Base Model Construction<br>Linear Regression, Decision Tree"]
  D --> F["Advanced Algorithm<br>LightGBM"]
  E --> G["Ensemble Learning Integration<br>Averaging, Voting, Blending, Stacking"]
  F --> G
  G --> H["Key Outcomes: <br>Optimized Fusion Model & <br>Improved Credit Evaluation Strategy"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a credit assessment model for mobile Internet users that fuses LightGBM with other ensemble methods to improve accuracy.

Source sentence
this paper uses the massive data provided by communication operators to carry out research on the operator's user credit evaluation model based on the fusion LightGBM algorithm.

Assumes 1

  • The selected feature subsets have a meaningful correlation with user credit scores. data
    Source sentence
    In order to better observe the correlation between different types of features and credit scores, manual segmentation is selected here.

Data

  • Consumer Capacity · communication operators alternative proprietary
    operator users
    Source sentence
    The consumer capacity dataset is mainly used to measure the user’s related characteristics in the mobile communication business.
  • Location Trajectory · communication operators alternative proprietary
    operator users
    Source sentence
    The location trajectory dataset is mainly used to measure the user’s daily activity location trajectory related characteristics.
  • Application Behavior Preference · communication operators alternative proprietary
    operator users
    Source sentence
    Application preference is mainly used to reflect the user’s application usage.

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • LightGBM achieves the best performance on the Consumer Capacity dataset with MAE 22.1867, MAPE 0.3701, MSE 935.7114, RMSE 30.5894 and R2 0.6534, outperforming LR, DT and RF. performance
    Source sentence
    LightGBM’s indicators are far better than the LR, DT, and RF basic models, proving that it does have good accuracy.
  • On the Location Trajectory dataset LightGBM yields the lowest errors (MAE 30.7300, MAPE 0.5144, MSE 1547.7564, RMSE 39.3415) and highest R2 (0.1464) among the four models. performance
    Source sentence
    LightGBM’s indicators are still slightly better than the LR, DT, and RF basic models.
  • For Application Behavior Preference LightGBM attains the best results (MAE 23.1702, MAPE 0.3881, MSE 973.7953, RMSE 31.2057, R2 0.4630) compared with the other algorithms. performance
    Source sentence
    LightGBM’s indicators are far better than the LR, DT, and RF basic models, proving that it does have good accuracy.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 29 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (14)

  1. Xgboost: A scalable tree boosting system (2016)
  2. Research on credit rating method based on BP NN (2007)
  3. International networks of banks and industry (2012)
  4. Bayesian credit scoring model with integration of expert knowledge and customer data (2010)
  5. Statistical model evaluation and information criteria (1999)
  6. Identification of a standard AI based technique for credit risk analysis (2016)
  7. Big data and supply chain decisions: the impact of volume, variety and velocity properties on the bullwhip effect (2017)
  8. Gradient boosted decision trees for high dimensional sparse output (2017)
and 6 more
  1. Lightgbm: A highly efficient gradient boosting decision tree (2017)
  2. Ensemble learning: A survey (2018)
  3. Financial credit risk assessment: a recent review (2016)
  4. Random forests (2001)
  5. Ensemble Methodology: Innovations in Credit Default Prediction Using LightGBM, XGBoost, and LocalEnsemble (2024) — link
  6. Enhancing Credit Card Fraud Detection: A Neural Network and SMOTE Integrated Approach (2024)

0 of 14 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.