New research has revealed the breakdown of the types of queries submitted to search engines.
In a study that has the potential to have an effect on search engine marketing techniques, researchers have revealed that some 80 per cent of search queries are for informational purposes, while ten per cent are navigational and the remaining ten per cent are for transactional reasons.
The study was the first of its kind using actual search data and aimed to achieve real-time classification.
In total, more than 1.5 million queries from hundreds of thousands of different search engine users were studied, with the results showing that the majority of queries were for informational purposes.
The researchers class informational searches as those that are looking for a specific piece of information, while navigational ones are where a specific website is sought and transactional ones are related to looking for a specifically piece of information related to buying a particular product.
Leader of the study Jim Jansen said that he and his colleagues developed an algorithm for the classification of searches with a 74 per cent rate of accuracy, noting that on the other queries, user intent is vague, meaning that probabilistic classification is needed.
He noted that the findings have a range of implications for both e-commerce and search engines if it is possible to classify the intent of users in real time.
Jim Jansen is the assistant professor at the College of Information Sciences and Technology in Penn State.
He said that tests on this topic that have been carried out before have classified much smaller sets of queries and that this has usually been done manually, whereas this study classified them automatically.
The researchers believe that such findings could be used in the drive to offer more targeted search results and improve search engine optimisation.