Analysis　of　Web　user　browser　behavior　characteristics　is　a　key　technology　in　the　domains,　such　as　Initiative　Web　Information　Retrieval　and　Information　Recommendation.　Taking　into　account　the　layout　of　Web　pages,　in　this　paper　we　constructed　a　user　browse　behavior　characteristic　(BC)　tree　based　on　the　browsing　history　of　Web　users,　and　then　established　a　new　approach　for　analyzing　Web　user　BC　trees.　This　method　delivers　us　the　interesting　topics　of　a　user,　the　extent　and　depth　of　these　topics,　as　well　as　the　explanation　of　the　frequency　of　accessing　hierarchic　block　paths　on　web　pages.　We　illustrated　the　efficiency　with　experiments,　and　demonstrated　that　the　proposed　approach　is　promising　in　Initiative　Web　Information　Retrieval　and　Information　Recommendation.　©　2010　Springer-Verlag.