open-domain question answering (lehnert, 1986; harabagiu et al, 2001; light et al, 2001) and storycomprehension (hirschman et al, 1999) have become important directions in natural language pro cessing. question answering is a retrieval task morechallenging than common search engine tasks be cause its purpose is to find an accurate and conciseanswer to a question rather than a relevant docu ment. the difficulty is more acute in tasks such as story comprehension in which the target text is less likely to overlap with the text in the questions. for this reason, advanced natural language techniques rather than simple key term extraction are needed.one of the important stages in this process is analyz ing the question to a degree that allows determining the ?type? of the sought after answer. in the treccompetition (voorhees, 2000), participants are requested to build a system which, given a set of en glish questions, can automatically extract answers (a short phrase) of no more than 50 bytes from a5-gigabyte document library. participants have re research supported by nsf grants iis-9801638 and itr iis 0085836 and an onr muri award. alized that locating an answer accurately hinges on first filtering out a wide range of candidates (hovy et al, 2001; ittycheriah et al, 2001) based on some categorization of answer types. this work develops a machine learning approach to question classification (qc) (harabagiu et al, 2001; hermjakob, 2001). our goal is to categorize questions into different semantic classes that impose constraints on potential answers, so that they can be utilized in later stages of the question answeringprocess. for example, when considering the question q: what canadian city has the largest popula tion?, the hope is to classify this question as havinganswer type city, implying that only candidate an swers that are cities need consideration.based on the snow learning architecture, we develop a hierarchical classifier that is guided by a lay ered semantic hierarchy of answer types and is able to classify questions into fine-grained classes. wesuggest that it is useful to consider this classifica tion task as a multi-label classification and find that it is possible to achieve good classification results(over 90%) despite the fact that the number of dif ferent labels used is fairly large, 50. we observe thatlocal features are not sufficient to support this accu racy, and that inducing semantic features is crucial for good performance. the paper is organized as follows: sec. 2 presents the question classification problem; sec. 3 discusses the learning issues involved in qc and presents ourlearning approach; sec. 4 describes our experimen tal study.this paper presents a machine learning approach to question classification. 4 describes our experimen tal study. in future work we plan to investigate further the application of deeper semantic analysis (including better named entity and semantic categorization) to feature extraction, automate the generation of thesemantic features and develop a better understand ing to some of the learning issues involved in thedifference between a flat and a hierarchical classi fier. question answering is a retrieval task morechallenging than common search engine tasks be cause its purpose is to find an accurate and conciseanswer to a question rather than a relevant docu ment. we define question classification(qc) here to be the task that, given a question, maps it to one of k classes, which provide a semantic constraint on the sought-after answer1. open-domain question answering (lehnert, 1986; harabagiu et al, 2001; light et al, 2001) and storycomprehension (hirschman et al, 1999) have become important directions in natural language pro cessing. the ambiguity causes the classifier not to output equivalent term as the first choice. we designed two experiments to test the accuracy ofour classifier on trec questions. what do bats eat?. in this case, both classes are ac ceptable. the first experi ment evaluates the contribution of different featuretypes to the quality of the classification.