Web query classification

Last updated

A Web query topic classification/categorization is a problem in information science. The task is to assign a Web search query to one or more predefined categories, based on its topics. The importance of query classification is underscored by many services provided by Web search. A direct application is to provide better search result pages for users with interests of different categories. For example, the users issuing a Web query "apple" might expect to see Web pages related to the fruit apple, or they may prefer to see products or news related to the computer company. Online advertisement services can rely on the query classification results to promote different products more accurately. Search result pages can be grouped according to the categories predicted by a query classification algorithm. However, the computation of query classification is non-trivial. Different from the document classification tasks, queries submitted by Web search users are usually short and ambiguous; also the meanings of the queries are evolving over time. Therefore, query topic classification is much more difficult than traditional document classification tasks.

Contents

Difficulties

Web query topic classification is to automatically assign a query to some predefined categories. Different from the traditional document classification tasks, there are several major difficulties which hinder the progress of Web query understanding:

Derive an appropriate feature representation for Web queries

Many queries are short and query terms are noisy. As an example, in the KDDCUP 2005 dataset, queries containing 3 words are most frequent (22%). Furthermore, 79% queries have no more than 4 words. A user query often has multiple meanings. For example, "apple" can mean a kind of fruit or a computer company. "Java" can mean a programming language or an island in Indonesia. In the KDDCUP 2005 dataset, most of the queries contain more than one meaning. Therefore, only using the keywords of the query to set up a vector space model for classification is not appropriate.

Query-enrichment based methods [1] [2] start by enriching user queries to a collection of text documents through search engines. Thus, each query is represented by a pseudo-document which consists of the snippets of top ranked result pages retrieved by search engine. Subsequently, the text documents are classified into the target categories using synonym based classifier or statistical classifiers, such as Naive Bayes (NB) and Support Vector Machines (SVMs).

Adapting to changes of the queries and categories over time

The meanings of queries may also evolve over time. Therefore, the old labeled training queries may be out-of-data and useless soon. How to make the classifier adaptive over time becomes a big issue. For example, the word "Barcelona" has a new meaning of the new micro-processor of AMD, while it refers to a city or football club before 2007. The distribution of the meanings of this term is therefore a function of time on the Web.

Intermediate taxonomy based method [3] first builds a bridging classifier on an intermediate taxonomy, such as Open Directory Project (ODP), in an offline mode. This classifier is then used in an online mode to map user queries to the target categories via the intermediate taxonomy. The advantage of this approach is that the bridging classifier needs to be trained only once and is adaptive for each new set of target categories and incoming queries.

Using unlabeled query logs to help with query classification

Since the manually labeled training data for query classification is expensive, how to use a very large web search engine query log as a source of unlabeled data to aid in automatic query classification becomes a hot issue. These logs record the Web users' behavior when they search for information via a search engine. Over the years, query logs have become a rich resource which contains Web users' knowledge about the World Wide Web.

Query clustering method [4] tries to associate related queries by clustering "session data", which contain multiple queries and click-through information from a single user interaction. They take into account terms from result documents that a set of queries has in common. The use of query keywords together with session data is shown to be the most effective method of performing query clustering.

Selectional preference based method [5] tries to exploit some association rules between the query terms to help with the query classification. Given the training data, they exploit several classification approaches including exact-match using labeled data, N-Gram match using labeled data and classifiers based on perception. They emphasize on an approach adapted from computational linguistics named selectional preferences. If x and y form a pair (x; y) and y belongs to category c, then all other pairs (x; z) headed by x belong to c. They use unlabeled query log data to mine these rules and validate the effectiveness of their approaches on some labeled queries.

Applications

All these services rely on the understanding Web users' search intents through their Web queries.

See also

Related Research Articles

Information retrieval (IR) in computing and information science is the task of identifying and retrieving information system resources that are relevant to an information need. The information need can be specified in the form of a search query. In the case of document retrieval, queries can be based on full-text or other content-based indexing. Information retrieval is the science of searching for information in a document, searching for documents themselves, and also searching for the metadata that describes data, and for databases of texts, images or sounds.

In computing, a search engine is an information retrieval software system designed to help find information stored on one or more computer systems. Search engines discover, crawl, transform, and store information for retrieval and presentation in response to user queries. The search results are usually presented in a list and are commonly called hits. The most widely used type of search engine is a web search engine, which searches for information on the World Wide Web.

An image retrieval system is a computer system used for browsing, searching and retrieving images from a large database of digital images. Most traditional and common methods of image retrieval utilize some method of adding metadata such as captioning, keywords, title or descriptions to the images so that retrieval can be performed over the annotation words. Manual image annotation is time-consuming, laborious and expensive; to address this, there has been a large amount of research done on automatic image annotation. Additionally, the increase in social web applications and the semantic web have inspired the development of several web-based image annotation tools.

Document classification or document categorization is a problem in library science, information science and computer science. The task is to assign a document to one or more classes or categories. This may be done "manually" or algorithmically. The intellectual classification of documents has mostly been the province of library science, while the algorithmic classification of documents is mainly in information science and computer science. The problems are overlapping, however, and there is therefore interdisciplinary research on document classification.

When classification is performed by a computer, statistical methods are normally used to develop the algorithm.

Search engine indexing is the collecting, parsing, and storing of data to facilitate fast and accurate information retrieval. Index design incorporates interdisciplinary concepts from linguistics, cognitive psychology, mathematics, informatics, and computer science. An alternate name for the process, in the context of search engines designed to find web pages on the Internet, is web indexing.

A web query or web search query is a query that a user enters into a web search engine to satisfy their information needs. Web search queries are distinctive in that they are often plain text and boolean search directives are rarely used. They vary greatly from standard query languages, which are governed by strict syntax rules as command languages with keyword or positional parameters.

Human–computer information retrieval (HCIR) is the study and engineering of information retrieval techniques that bring human intelligence into the search process. It combines the fields of human-computer interaction (HCI) and information retrieval (IR) and creates systems that improve search by taking into account the human context, or through a multi-step search process that provides the opportunity for human feedback.

Document clustering is the application of cluster analysis to textual documents. It has applications in automatic document organization, topic extraction and fast information retrieval or filtering.

A concept search is an automated information retrieval method that is used to search electronically stored unstructured text for information that is conceptually similar to the information provided in a search query. In other words, the ideas expressed in the information retrieved in response to a concept search query are relevant to the ideas contained in the text of the query.

In machine learning, one-class classification (OCC), also known as unary classification or class-modelling, tries to identify objects of a specific class amongst all objects, by primarily learning from a training set containing only the objects of that class, although there exist variants of one-class classifiers where counter-examples are used to further refine the classification boundary. This is different from and more difficult than the traditional classification problem, which tries to distinguish between two or more classes with the training set containing objects from all the classes. Examples include the monitoring of helicopter gearboxes, motor failure prediction, or the operational status of a nuclear plant as 'normal': In this scenario, there are few, if any, examples of catastrophic system states; only the statistics of normal operation are known.

DeepPeep was a search engine that aimed to crawl and index every database on the public Web. Unlike traditional search engines, which crawl existing webpages and their hyperlinks, DeepPeep aimed to allow access to the so-called Deep web, World Wide Web content only available via for instance typed queries into databases. The project started at the University of Utah and was overseen by Juliana Freire, an associate professor at the university's School of Computing WebDB group. The goal was to make 90% of all WWW content accessible, according to Freire. The project ran a beta search engine and was sponsored by the University of Utah and a $243,000 grant from the National Science Foundation. It generated worldwide interest.

Yebol was a vertical "decision" search engine that had developed a knowledge-based, semantic search platform. Based in San Jose, California, Yebol's artificial intelligence human intelligence-infused algorithms automatically cluster and categorize search results, web sites, pages and contents that it presents in a visually indexed format that is more aligned with initial human intent. Yebol used association, ranking and clustering algorithms to analyze related keywords or web pages. Yebol presented as one of its goals the creation of a unique "homepage look" for every possible search term.

Learning to rank or machine-learned ranking (MLR) is the application of machine learning, typically supervised, semi-supervised or reinforcement learning, in the construction of ranking models for information retrieval systems. Training data may, for example, consist of lists of items with some partial order specified between items in each list. This order is typically induced by giving a numerical or ordinal score or a binary judgment for each item. The goal of constructing the ranking model is to rank new, unseen lists in a similar way to rankings in the training data.

International Aging Research Portfolio (IARP) is a non-profit, open-access knowledge management system incorporating grants, publications, conferences in natural and social & behavioral sciences. In addition to the advanced search and visual trend analysis tools the system includes a directory of research projects classified into categories related to aging research. The system uses automatic classification algorithms with elements of machine learning to assign research projects to the relevant categories. The directory is curated by many expert category editors and science advisory board members. The chair of the science advisory board is Dr. Charles Cantor.

In web analytics, a session, or visit is a unit of measurement of a user's actions taken within a period of time or with regard to completion of a task. Sessions are also used in operational analytics and provision of user-specific recommendations. There are two primary methods used to define a session: time-oriented approaches based on continuity in user activity and navigation-based approaches based on continuity in a chain of requested pages.

Automatic taxonomy construction (ATC) is the use of software programs to generate taxonomical classifications from a body of texts called a corpus. ATC is a branch of natural language processing, which in turn is a branch of artificial intelligence.

Online content analysis or online textual analysis refers to a collection of research techniques used to describe and make inferences about online material through systematic coding and interpretation. Online content analysis is a form of content analysis for analysis of Internet-based communication.

User intent, otherwise known as query intent or search intent, is the identification and categorization of what a user online intended or wanted to find when they typed their search terms into an online web search engine for the purpose of search engine optimisation or conversion rate optimisation. Examples of user intent are fact-checking, comparison shopping or navigating to other websites.

Weak supervision is a paradigm in machine learning, the relevance and notability of which increased with the advent of large language models due to large amount of data required to train them. It is characterized by using a combination of a small amount of human-labeled data, followed by a large amount of unlabeled data. In other words, the desired output values are provided only for a subset of the training data. The remaining data is unlabeled or imprecisely labeled. Intuitively, it can be seen as an exam and labeled data as sample problems that the teacher solves for the class as an aid in solving another set of problems. In the transductive setting, these unsolved problems act as exam questions. In the inductive setting, they become practice problems of the sort that will make up the exam. Technically, it could be viewed as performing clustering and then labeling the clusters with the labeled data, pushing the decision boundary away from high-density regions, or learning an underlying one-dimensional manifold where the data reside.

References

  1. ↑ Shen et al. "Q2C@UST: Our Winning Solution to Query Classification". ACM SIGKDD Exploration, December 2005, Volume 7, Issue 2.
  2. ↑ Shen et al. "Query Enrichment for Web-query Classification". ACM TOIS, Vol. 24, No. 3, July 2006.
  3. ↑ Shen et al. "Building bridges for web query classification". ACM SIGIR, 2006.
  4. ↑ Wen et al. "Query Clustering Using User Logs", ACM TOIS, Volume 20, Issue 1, January 2002.
  5. ↑ Beitzel et al. "Automatic Classification of Web Queries Using Very Large Unlabeled Query Logs", ACM TOIS, Volume 25, Issue 2, April 2007.
  6. ↑ Data Mining and Audience Intelligence for Advertising (ADKDD'07), KDD workshop 2007
  7. ↑ Targeting and Ranking for Online Advertising (TROA'08), WWW workshop 2008

Further reading

This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.