The sequence pattern mining method aims to identify frequent sequences that exceed a user-specified support threshold. The present study uses the same approach based on sequential standards to estimate the heat stress of broilers from a resulting behavioural pattern. Experimental data were recorded in a climate chamber where the behaviour of broilers was recorded under thermoneutral (comfort) conditions, set as standard, and when exposed to thermal stress (cold and heat). The Generalised Sequential Patterns (GSP) algorithm was used to evaluate the heat stress of broilers in the third and fourth week of growth. The results indicated that the mining of pattern sequences is a useful and straightforward technique to estimate the welfare of broiler chickens, allowing the identification of temporal relations between thermal stress and the consequent behaviour of the broiler. Temperature 8 °C below the standard thermoneutral conditions showed that the broiler remained lying down most of the time, walking only to the drinker and feeder trough. Broilers exposed to temperatures 8 °C above the standard thermoneutral conditions () tend to decrease locomotor activities, showing lower welfare status.

中文翻译：

---

用广义序列模式算法检测肉鸡热应激

序列模式挖掘方法旨在识别超过用户指定支持阈值的频繁序列。本研究使用基于顺序标准的相同方法从由此产生的行为模式估计肉鸡的热应激。实验数据记录在气候室中，其中在热中性（舒适）条件下记录肉鸡的行为，设置为标准，以及暴露于热应力（冷和热）时。使用广义序列模式（GSP）算法评估肉鸡在生长第三周和第四周的热应激。结果表明，模式序列的挖掘是一种有用且直接的技术来估计肉鸡的福利，允许识别热应力与肉鸡随后的行为之间的时间关系。低于标准热中性条件 8 °C 的温度表明，肉鸡大部分时间都保持躺着状态，仅步行到饮水器和饲养槽。肉鸡暴露在比标准热中性条件 () 高 8 °C 的环境中，往往会降低运动活动，表现出较低的福利状况。