In this paper, we propose a semi-supervised learning approach for classifying program (bot) generated web search traffic from that of genuine human users. The work is motivated by the challenge that the enormous amount of search data pose to traditional approaches that rely on fully annotated training samples. We propose a semi-supervised framework that addresses the problem in multiple fronts. First, we use the CAPTCHA technique and simple heuristics to extract from the data logs a large set of training samples with initial labels, though directly using these training data is problematic because the data thus sampled are biased. To tackle this problem, we further develop a semi-supervised learning algorithm to take advantage of the unlabeled data to improve the classification performance. These two proposed algorithms can be seamlessly combined and very cost efficient to scale the training process. In our experiment, the proposed approach showed significant (i.e. 2:1) improvement compared to the traditional supervised approach.

In this article, we report our efforts in mining the information encoded as clickthrough data in the server logs to evaluate and monitor the relevance ranking quality of a commercial web search engine. We describe a metric called pSkip that aims to quantify the ranking quality by estimating the probability of users encountering non relevant results that cost them the efforts to read and skip. A search engine with a lower pSkip is regarded as having a better ranking quality. A key design goal of pSkip is to integrate the findings from two sets of user studies that utilize eye-tracking devices to track users' browsing patterns on the search result pages, and that use specially instrumented browsers to actively solicit users' explicit judgments on their search activities. We present the derivation of the maximum likelihood estimation of pSkip and demonstrate its efficacy in describing the user study data. The mathematical properties of pSkip are further analyzed and compared with several objective metrics as well as the cumulated gain method that uses subjective judgments. Experimental data show that pSkip can measure aspects of the search quality that these existing metrics are not designed or fail to address, such as identifying the real search intents expressed in the ambiguous queries. Although effective and superior in many ways, we also report a series of experiments that show pSkip may be influenced by system issues that are not directly related to relevance ranking, suggesting that measurements complementary to pSkip are still needed in order to form a holistic and accurate characterization of the ranking quality.

Personal names are an important kind of Web queries in Web search, and yet they are special in many ways. Strategies for retrieving information on personal names should therefore be different from the strategies for other types of queries. To improve the search quality for personal names, a first step is to detect whether a query is a personal name. Despite the importance of this problem, relatively little previous research has been done on this topic. Since Web queries are usually short, conventional supervised machine-learning algorithms cannot be applied directly. An alternative is to apply some heuristic rules coupled with name-term dictionaries. However, when the dictionaries are small, this method tends to make false negatives; when the dictionaries are large, it tends to generate false positives. A more serious problem is that this method cannot provide a good trade-off between precision and recall. To solve these problems, we propose an approach based on the construction of probabilistic name-term dictionaries and personal name grammars, and use this algorithm to predict the probability of a query to be a personal name. In this paper, we develop four different methods for building probabilistic name-term dictionaries in which a term is assigned with a probability value of the term being a name term. We compared our approach with baseline algorithms such as dictionary-based look-up methods and supervised classification algorithms including logistic regression and SVM on some manually labeled test sets. The results validate the effectiveness of our approach, whose F1 value is more than 79.8%, which outperforms the best baseline by more than 11.3%

The top web search result is crucial for user satisfaction with the web search experience. We argue that the importance of the relevance at the top position necessitates special handling of the top web search result for some queries. We propose an effective approach of leveraging millions of past user interactions with a web search engine to automatically detect "best bet" top results preferred by majority of users. Interestingly, this problem can be more effectively addressed with classification than using state-of-the-art general ranking methods. Furthermore, we show that our general machine learning approach achieves precision comparable to a heavily tuned, domain-specific algorithm, with significantly higher coverage. Our experiments over millions of user interactions for thousands of queries demonstrate the effectiveness and robustness of our techniques.

Ensemble learning strategies, especially boosting and bagging decision trees, have demonstrated impressive capacities to improve the prediction accuracy of base learning algorithms. Further gains have been demonstrated by strategies that combine simple ensemble formation approaches. We investigate the hypothesis that the improvement in accuracy of multistrategy approaches to ensemble learning is due to an increase in the diversity of ensemble members that are formed. In addition, guided by this hypothesis, we develop three new multistrategy ensemble learning techniques. Experimental results in a wide variety of natural domains suggest that these multistrategy ensemble learning techniques are, on average, more accurate than their component ensemble learning techniques.

This article investigates boosting naive Bayesian classification. It first shows that boosting does not improve the accuracy of the naive Bayesian classifier as much as we expected in a set of natural domains. By analyzing the reason for boosting's weakness, we propose to introduce tree structures into naive Bayesian classification to improve the performance of boosting when working with naive Bayesian classification. The experimental results show that although introducing tree structures into naive Bayesian classification increases the average error of the naive Bayesian classification for individual models, boosting naive Bayesian classifiers with tree structures can achieve significantly lower average error than both the naive Bayesian classifier and boosting the naive Bayesian classifier, providing a method of successfully applying the boosting technique to naive Bayesian classification. A bias and variance analysis confirms our expectation that the naive Bayesian classifier is a stable classifier with low variance and high bias. We show that the boosted naive Bayesian classifier has a strong bias on a linear form, exactly the same as its base learner. Introducing tree structures reduces the bias and increases the variance, and this allows boosting to gain advantage.

We show that the e-commerce domain can provide all the right ingredients for successful data mining. We describe an integrated architecture for supporting this integration. The architecture can dramatically reduce the pre-processing, cleaning, and data understanding effort often documented to take 80% of the time in knowledge discovery projects. We emphasize the need for data collection at the application server layer (not the Web server) in order to support logging of data and metadata that is essential to the discovery process. We describe the data transformation bridges required from the transaction processing systems and customer event streams (e.g., clickstreams) to the data warehouse. We detail the mining workbench, which needs to provide multiple views of the data through reporting, data mining algorithms, visualization, and OLAP. We conclude with a set of challenges.

This study compares five well-known association rule algorithms using three real-world datasets and an artificial dataset. The experimental results confirm the performance improvements previously claimed by the authors on the artificial data, but some of these gains do not carry over to the real datasets, indicating overfitting of the algorithms to the IBM artificial dataset. More importantly, we found that the choice of algorithm only matters at support levels that generate more rules than would be useful in practice. For support levels that generate less than 1,000,000 rules, which is much more than humans can handle and is sufficient for prediction purposes where data is loaded into RAM, Apriori finishes processing in less than 10 minutes. On our datasets, we observed super-exponential growth in the number of rules. On one of our datasets, a 0.02% change in the support increased the number of rules from less than a million to over a billion, implying that outside a very narrow range of support values, the choice of algorithm is irrelevant.

Techniques for constructing classifier committees including boosting and bagging have demonstrated great success, especially boosting for decision tree learning. This type of technique generates several classifiers to form a committee by repeated application of a single base learning algorithm. The committee members vote to decide the final classification. Boosting and bagging create different classifiers by modifying the distribution of the training set. SASC (Stochastic Attribute Selection Committees) uses an alternative approach to generating classifier committees by stochastic manipulation of the set of attributes considered at each node during tree induction, but keeping the distribution of the training set unchanged. We propose a method for improving the performance of boosting. This technique combines boosting and SASC. It builds classifier committees by manipulating both the distribution of the training set and the set of attributes available during induction. In the synergy SASC effectively increases the model diversity of boosting. Experiments with a representative collection of natural domains show that, on average, the combined technique outperforms either boosting or SASC alone in terms of reducing the error rate of decision tree learning.