# US-Rule: Discovering Utility-driven Sequential Rules

arXiv: https://arxiv.org/abs/2111.15020 · HTML: https://mlsanity.com/paper/2111.15020/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 43 references that are in the map.
- First posted on arXiv 29 Nov 2021.

## Builds on (most cited first, 6 of 6)

- [A Survey of Utility-Oriented Pattern Mining](https://mlsanity.com/paper/1805.10511/) (2018)
- [A Survey of Parallel Sequential Pattern Mining](https://mlsanity.com/paper/1805.10515/) (2018)
- [ProUM: Projection-based Utility Mining on Sequence Data](https://mlsanity.com/paper/1904.07764/) (2019)
- [TUSQ: Targeted High-Utility Sequence Querying](https://mlsanity.com/paper/2103.16615/) (2021)
- [TKUS: Mining Top-K High-Utility Sequential Patterns](https://mlsanity.com/paper/2011.13454/) (2020)
- [On-shelf Utility Mining of Sequence Data](https://mlsanity.com/paper/2011.13455/) (2020)

## Built on by (most cited first, 6 of 6)

- [Large Language Models for Medicine: A Survey](https://mlsanity.com/paper/2405.13055/) (2024)
- [Large Language Models for Education: A Survey](https://mlsanity.com/paper/2405.13001/) (2024)
- [Towards Target Sequential Rules](https://mlsanity.com/paper/2206.04728/) (2022)
- [Totally-ordered Sequential Rules for Utility Maximization](https://mlsanity.com/paper/2209.13501/) (2022)
- [MDL-based Compressing Sequential Rules](https://mlsanity.com/paper/2212.10252/) (2022)
- [Discovering Utility-driven Interval Rules](https://mlsanity.com/paper/2309.16102/) (2023)

## Abstract (arXiv)

Utility-driven mining is an important task in data science and has many applications in real life. High utility sequential pattern mining (HUSPM) is one kind of utility-driven mining. HUSPM aims to discover all sequential patterns with high utility. However, the existing algorithms of HUSPM can not provide an accurate probability to deal with some scenarios for prediction or recommendation. High-utility sequential rule mining (HUSRM) was proposed to discover all sequential rules with high utility and high confidence. There is only one algorithm proposed for HUSRM, which is not enough efficient. In this paper, we propose a faster algorithm, called US-Rule, to efficiently mine high-utility sequential rules. It utilizes rule estimated utility co-occurrence pruning strategy (REUCP) to avoid meaningless computation. To improve the efficiency on dense and long sequence datasets, four tighter upper bounds (LEEU, REEU, LERSU, RERSU) and their corresponding pruning strategies (LEEUP, REEUP, LERSUP, RERSUP) are proposed. Besides, US-Rule proposes rule estimated utility recomputing pruning strategy (REURP) to deal with sparse datasets. At last, a large number of experiments on different datasets compared to the state-of-the-art algorithm demonstrate that US-Rule can achieve better performance in terms of execution time, memory consumption and scalability.