In　this　paper,　we　used　semi-Conditional　Random　Fields　(semi-CRFs)　model,　a　conditionally　trained　version　of　semi-Markov　chains　for　sequence　labeling.　It　is　a　valid　probabilistic　model　to　segment　and　label　sequence　data.　The　key　advantage　of　semi-CRFs　over　hidden　Markov　models　is　their　conditional　nature,　resulting　in　the　relaxation　of　the　independence　assumptions　required　by　HMMs　in　order　to　ensure　tractable　inference.　Additionally,　semi-CRFs　avoid　the　label　bias　problem,　a　weakness　exhibited　by　maximum　entropy　Markov　models　and　other　conditional　Markov　models　based　on　directed　graphical　models.　Experimental　results　show　that　the　semi-CRFs　outperform　on　real-world　sequence　labeling　tasks.　Copyright　©　2010　Binary　Information　Press　May,　2010.