A search engine in your browser

by Djoerd Hiemstra

Information Retrieval has come a long way since Cyril Cleverdon conducted his famous Cranfield experiments on 1400 paper abstracts in the 1960’s. At the time, searching the massive 1.6 megabyte Cranfield collection was a big deal. Today, we routinely take pictures that are much bigger, and even a crappy smart... [Read More]
Tags: privacy

A university's autocompletion ABC

Last week we launched our software for query autocompletions: Type in the first character of your query, and Searsia will recommend a query that best completes your prefix. The autocompletions are targeted to the users of the search engine, without the need to track those users. Searsia’s autocompletions are demonstrated... [Read More]
Tags: autocompletions

Query suggestions without tracking users

by Djoerd Hiemstra

Today, we launch searsiasuggest, a simple but effective open source solution for query suggestions. It implements 3 kinds of query suggestions: 1) autocompletions, 2) spelling correction, and 3) related queries. The software works pretty much like many similar solutions: It inputs a list of queries with their popularity, it indexes... [Read More]
Tags: autocompletions

Query autocompletions considered harmful

by Djoerd Hiemstra

Search engines suggest completions of partially typed queries to help users speed up their searches, for instance by showing the suggested completions in a drop down box. These query autocompletions enable the user to search faster, searching for long queries with relatively few key strokes. Jakobsson (1986) showed that for... [Read More]
Tags: autocompletions

Searsia Cartoon by Anne Stalinski

The Dutch chapter of ISOC asked the great Anne Stalinksi to make a cartoon of Searsia. Translated to English, the cartoon says: “Searsia allows users to search sources that Google cannot use”. The user enters the query “How to bake a quiche?”, and Searsia not only searches the “Internet”, but... [Read More]
Tags: isoc