Author Archives: LiqiangGuo

我胡汉三又回到wordpress啦 哈哈 多年前，在对学术对科研无比热烈的岁月中，create了word process，之后被GFW成功阻拦在了围墙之内， 现在天天忙于给资本家码代码，一时兴起 也许是为了满足虚荣欲，在google scholar search了一把自己的名字，于是我们又见面了，多年不见，你可还好 word press！ 五年已过，我已是另一番模样。

Selecting Good Expansion Terms for Pseudo-Relevance Feedback. But I think proximity is too simple,maybe some advanced approach should be studied, for example topic model. RBF is useful in IR，but because of the existence of multi-topics in a document (some may … Continue reading →

Latent Dirichlet Allocation Probabilistic Topic Models Latent Dirichlet Allocation2

Probabilistic Latent Semantic Indexing Unsupervised Learning by Probabilistic Latent Semantic Analysis

The LSA is unable to explicitly capture multiple senses of a word, nor does it take into account that every word occurrence is typically intended to refer to only one meaning at a time. So the LSA can not address … Continue reading →

Gerative Model andd Discriminative Model

Making tree kernels practical for natural language learning

Note:When the accounts of the returned dcoument lists belonging to different queries are the same, Macro==Micro. Macro_Micro

Zipf’s Law

1.Relevance Model 2.Divergence minimizatioon model 3.simple mixture model 4.regularied mixture model A Comparative Study of Methods for Estimating Query Language Models with Pseudo Feedback