We describe an algorithm that relies on web frequency counts to identify and correct writing errors made by non-native writers of English. Evaluation of the system on a realworld ESL corpus showed very promising performance on the very difficult problem of critiquing English determiner use: 62% precision and 41% recall, with a false flag rate of only 2% (compared to a random-guessing baseline o...

We evaluate the output of a Markov model-based harmonic progression generator, a classic model for corpus-based computational creativity. 87 participants performed a discrimination task classifying 20 musical excerpts as either human-composed or computer-composed. Also recorded was each participant's level of confidence in their choice. Results indicated that while overall performance was above...

Behavioral game theory models are important in organizing experimental data of strategic decision making. However, are subjects classified as behavioral types more predictable than unclassified subjects? Alternatively, how many predictable subjects await new behavioral models to describe them? In our experiments, subjects play simple guessing games against random opponents and are subsequently ...

Behavioral game theory models are important in organizing experimental data of strategic decision making. However, are subjects classified as behavioral types more predictable in their choices than unclassified subjects? Alternatively, how many subjects await new behavioral models to describe them? In our experiments, subjects play simple guessing games against random opponents and are subseque...

A deck of cards consisting of m red and n black cards is given. A guess is made as to the colour of the top card, after which it is revealed and discarded. To maximise the number of correct guesses one chooses the colour corresponding to the majority of cards remaining in the deck. We determine the expected number of correct guesses with this strategy, as well as the probability of beating an o...

An experiment in beer tasting among 18 first year pre-clinical students showed that only two appeared to be able to discriminate light from bitter ale correctly. As a whole the group did worse than would be expected by guessing at random on each tasting trial. Some psychological biases, one possibly as a result of the design, are indicated by the results and a possible remedy suggested. The dan...

However, suppose a learning algorithm cannot attain error rates less than some fixed amount, say 40%. Is it possible, despite this limitation, to still drive the error rate arbitrarily close to zero? This question motivates the weak learning model, which will be described below. This model will consider circumstances in which the learning algorithms can only do slightly better than random guess...

We present simple formal models explaining unusual properties of several biological classification tasks as follows. For these datasets the whole range of supervised learning techniques generate predictive models which classify independent test samples systematically below the performance of random guessing (hence the name anti-learning). We show that explanation of such “counter-intuitive” sup...