07 Prefix-Adaptive and Time-Sensitive Personalized Query Auto Completion

1.题目和关键词
Title:
Prefix-Adaptive and Time-Sensitive Personalized Query Auto Completion
前缀自适应和时间敏感的个性化查询自动完成
Keywords:
Query auto completion查询自动完成;
personalization个性化;
time-sensitive时间敏感;
long-tail长尾;
web search网络搜索.

2.摘要
—Query auto completion (QAC) methods recommend queries to search engine users when they start entering a query. Current QAC methods mostly rank query completions based on their past popularity, i.e., on the number of times they have previously been submitted as a query. However, query popularity changes over time and may vary drastically across users. Accordingly, the ranking of query completions should be adjusted. Previous time-sensitive and user-specific QAC methods have been developed
separately, yielding significant improvements over methods that are neither time-sensitive nor personalized. We propose a hybrid QAC method that is both time-sensitive and personalized. We extend it to handle long-tail prefixes, which we achieve by assigning optimal weights to the contribution from time-sensitivity and personalization. Using real-world search log datasets, we return top N query suggestions ranked by predicted popularity as estimated from popularity trends and cyclic popularity behavior; we rerank them by integrating similarities to a user’s previous queries (both in the current session and in previous sessions). Our method outperforms state-of-the-art time-sensitive QAC baselines, achieving total improvements of between 3 and 7 percent in terms of mean reciprocal rank (MRR). After optimizing the weights, our extended model achieves MRR improvements of between 4 and 8 percent.

当搜索引擎用户开始输入查询时,查询自动完成(QAC)方法会向用户推荐查询。当前的QAC方法主要根据其过去的流行程度(即先前已作为查询提交的次数)对查询完成情况进行排名。然而,查询的流行度会随着时间的推移而变化,并且可能会因用户而异。因此,应该调整查询完成的排序。以前的时间敏感和用户特定的QAC方法是分开开发的,与既不具有时间敏感也不具有个性化的方法相比,产生了显著的改进。我们提出了一种时间敏感和个性化的混合QAC方法。我们将其扩展为处理长尾前缀,通过将最佳权重分配给时间敏感和个性化的贡献来实现的。使用真实的搜索日志数据集,根据流行趋势和周期性流行行为,返回按预测流行度排序的前N个查询建议;我们通过整合与用户先前查询(在当前会话和以前会话中)的相似度来重新排列它们。我们的方法优于最先进的时间敏感QAC基线,在平均倒数排名(MRR)方面实现了3%至7%的总体改进。在优化权重后,我们的扩展模型实现了4%到8%的MRR改进。

3.创新点、学术价值
We address the challenge of query auto completion in a novel way by considering both time-sensitive query popularity and user-specific context.
(1)通过考虑时间敏感的查询流行性和用户特定的上下文来解决查询自动完成的挑战。
We propose a new query popularity prediction method that explores the cyclic behavior and recent trend of query popularity.
(2)提出了一种新的查询流行度预测方法,该方法可以探索查询流行度的周期性行为和近期趋势。
We extend our hybrid QAC model to deal with long-tail prefixes by optimizing the contributions from query popularity and user-specific context.
(3)通过优化查询流行度和用户特定上下文的贡献,该论文扩展了混合QAC模型来处理长尾前缀。

4.对结论的理解和对学习工作的启发
以往的有关查询自动完成的大多数工作都集中在时间敏感的最大似然估计或上下文感知的相似性上。在该论文中,同时结合了QAC问题的两个方面。为了解用户的搜索意图,通过个性化QAC扩展了时间敏感QAC方法,这可以推断出当前的搜索会话和以前的搜索任务中当前查询和先前查询在字符上的相似性。此外,该论文还针对长尾前缀对模型进行了调整。并在检查长尾前缀的流行度后为其分配最佳权重已进行优化,而不是使用固定权重。

未来研究工作:
对于将来的工作,我们打算仔细研究N>10的基于流行度的排名方法返回的前N个查询完成:我们能从排名在较低的查询完成情况中获得多少好处? 此外,我们的目标是将我们的方法转移到具有长期查询日志的其他数据集,这应该有助于从比我们当前工作中使用的AOL和SnV日志访问时间更长的周期性查询中受益。使QAC结果多样化在多大程度上是有利的?此外,我们可以考虑对提案中的所有前缀使用最佳权重,并通过使用在训练期的一组类似的用户日志来解决无法获取用户长期搜索日志导致的冷启动问题。另一个可能的步骤是为活跃用户,特别是专业搜索者建立个性化的时间模式,这需要从实际的查询术语概括到主题或意图。这可能有助于为其生成更好的查询完成度排名。

猜你喜欢

转载自blog.csdn.net/weixin_37996254/article/details/108901929