Previous work demonstrated that Web counts can be used to approximate bigram counts, suggesting that Web-based frequencies should be useful for a wide variety of Natural Language Processing (NLP) tasks. However, only a limited number of tasks have so far been tested using Web-scale data sets. The present article overcomes this limitation by systematically investigating the performance of Web-based models for several NLP tasks, covering both syntax and semantics, both generation and analysis, and a wider range of n-grams and parts of speech than have been previously explored. For the majority of our tasks, we find that simple, unsupervised models perform better when n-gram counts are obtained from the Web rather than from a large corpus. In some cases, performance can be improved further by using backoff or interpolation techniques that combine Web counts and corpus counts. However, unsupervised Web-based models generally fail to outperform supervised state-ofthe-art models trained on smaller corpora. We argue that Web-based models should therefore be used as a baseline for, rather than an alternative to, standard supervised models.

Previous work demonstrated that web counts can be used to approximate bigram frequencies, and thus should be useful for a wide variety of NLP tasks. So far, only two generation tasks (candidate selection for machine translation and confusion-set disambiguation) have been tested using web-scale data sets. The present paper investigates if these results generalize to tasks covering both syntax and semantics, both generation and analysis, and a larger range of n-grams. For the majority of tasks, we find that simple, unsupervised models perform better when n-gram frequencies are obtained from the web rather than from a large corpus. However, in most cases, web-based models fail to outperform more sophisticated state-of-theart models trained on small corpora. We argue that web-based models should therefore be used as a baseline for, rather than an alternative to, standard models. 1