Conducting a search directed by a hierarchy-free set of topics
A computer implemented method, and a software product for directing a search aided by a set of topics, the topics not necessarily having a hierarchy. Each topic is attached to at least one information item of a plurality of information items, such as a Web page identified by a URL. The method includes identifying one or more information items of the plurality of information items, and determining one or more suggested topics from the set of topics according to the attachments of the suggested topics to the identified information items, and providing search results to the user including at least some of the identified information items and at least one of the suggested topics. Selecting one of the suggested topics generates a new set of suggested topics, such that a hierarchy of topics is formed on the fly for a particular search.
1 . A computer implemented method of directing a search aided by a set of topics, the topics not necessarily having a hierarchy, each topic having at least one attachment to at least one information item of a plurality of information items, the method comprising:
identifying for a first searcher one or more information items of the plurality of information items according to search request information provided by the first searcher, the search request information including at least one of the group consisting of a search phrase and a subset of one or more search topics of the set of topics;
determining one or more suggested topics from the set of topics, the suggested topics being determined according to the attachments of the suggested topics to the one or more identified information items; and
providing search results to the first searcher, the search results including at least some of the identified information items and at least one of the suggested topics,
wherein the suggested topics include one or more refinement topics determined from the set of topics according to a refinement topic criterion using a refinement selection method,
such that the first searcher selecting one of the suggested topics generates a new set of suggested topics, and such that a hierarchy of topics is formed on the fly for a particular search for the first searcher.
2 . A method as recited in claim 1 , wherein at least some of the attachments between the topics and the information items are defined by users, a user being a searcher who has provided registration information.
3 . A method as recited in claim 1 , wherein a potential refinement topic, according to the refinement topic selection method, is a topic that has an attachment to at least one of the identified items, and wherein the refinement topics are selected from the potential refinement topics.
4 . A method as recited in claim 3 , wherein the refinement topic selection method includes a refinement topic ranking method to rank potential refinement topics to determine an ordering for the refinement topics.
5 . A method as recited in claim 4 , wherein the refinement topic ranking method determines a refinement measure, that, for each potential refinement topic, determines a refinement measure dependent on the number of identified information items that are in attachments to the potential refinement topic in relation to the number of identified information items.
6 . A method as recited in claim 5 , wherein the refinement ranking method includes:
for each potential refinement topic, determining a refinement measure as the number of identified information items that are in attachments to the potential refinement topic in relation to the number of identified information items,
such that the most highly ranked potential refinement topic is that whose refinement measure is closest to 50% expressed as a percentage, and the next to most highly ranked potential refinement topic is that whose refinement measure is next closest to 50% expressed as a percentage,
whereby a refinement topic whose refinement measure is 50% expressed as a percentage would split the identified information items into two equal-size sets of information items.
7 . A method as recited in claim 5 , wherein the refinement ranking method includes:
for each potential refinement topic, determining a refinement measure according to the sum of the squares of:
the number of identified information items that are in attachments to the potential refinement topic in relation to the number of identified information items, and
the number of identified information items that are not in attachments to the potential refinement topic in relation to the number of identified information items,
such that the most highly ranked potential refinement topic is that whose refinement measure is the smallest, and the next to most highly ranked potential refinement topic is that whose refinement measure is next smallest.
8 . A method as recited in claim 1 , wherein each information item is a Web page.
9 . A method as recited in claim 1 , wherein the identifying of one or more information items includes:
accepting the search request information from the first searcher; and
searching the plurality of information items according to the accepted search request information and to one or more search criteria to identify the one or more information items, the search criteria including that the identified information items each having an attachment to each topic in the accepted search request information.
10 . A method as recited in claim 9 , further comprising:
displaying the results of the searching to the first searcher.
11 . A carrier medium carrying one or more computer-readable code segments to instruct a processor of a processing system to execute a method of directing a search aided by a set of topics, the topics not necessarily having a hierarchy, each topic having at least one attachment to at least one information item of a plurality of information items, the method comprising:
identifying for a first searcher one or more information items of the plurality of information items according to search request information provided by the first searcher, the search request information including at least one of the group consisting of a search phrase and a subset of one or more search topics of the set of topics;
determining one or more suggested topics from the set of topics, the suggested topics being determined according to the attachments of the suggested topics to the one or more identified information items; and
providing search results to the first searcher, the search results including at least some of the identified information items and at least one of the suggested topics,
wherein the suggested topics include one or more refinement topics determined from the set of topics according to a refinement topic criterion using a refinement selection method,
such that the first searcher selecting one of the suggested topics generates a new set of suggested topics, and such that a hierarchy of topics is formed on the fly for a particular search for the first searcher.
12 . A method as recited in claim 11 , wherein at least some of the attachments between the topics and the information items are defined by users, a user being a searcher who has provided registration information.
13 . A carrier medium as recited in claim 11 , wherein a potential refinement topic according to the refinement topic selection method is a topic that has an attachment to at least one of the identified items, and wherein the refinement topics are selected from the potential refinement topics.
14 . A carrier medium as recited in claim 13 , wherein the refinement topic selection method includes a refinement topic ranking method to rank potential refinement topics to determine an ordering for the refinement topics.
15 . A carrier medium as recited in claim 14 , wherein the refinement topic ranking method determines a refinement measure, that, for each potential refinement topic, determines a refinement measure dependent on the number of identified information items that are in attachments to the potential refinement topic in relation to the number of identified information items.
16 . A carrier medium as recited in claim 15 , wherein the refinement ranking method includes:
for each potential refinement topic, determining a refinement measure as the number of identified information items that are in attachments to the potential refinement topic in relation to the number of identified information items,
such that the most highly ranked potential refinement topic is that whose refinement measure is closest to 50% expressed as a percentage, and the next to most highly ranked potential refinement topic is that whose refinement measure is next closest to 50% expressed as a percentage,
whereby a refinement topic whose refinement measure is 50% expressed as a percentage would split the identified information items into two equal-size sets of information items.
17 . A carrier medium as recited in claim 15 , wherein the refinement ranking method includes:
for each potential refinement topic, determining a refinement measure according to the sum of the squares of:
the number of identified information items that are in attachments to the potential refinement topic in relation to the number of identified information items, and
the number of identified information items that are not in attachments to the potential refinement topic in relation to the number of identified information items,
such that the most highly ranked potential refinement topic is that whose refinement measure is the smallest, and the next to most highly ranked potential refinement topic is that whose refinement measure is next smallest.
18 . A carrier medium as recited in claim 11 , wherein each information item is a Web page.
19 . A carrier medium as recited in claim 11 , wherein the identifying of one or more information items includes:
accepting the search request information from the first searcher; and
searching the plurality of information items according to the accepted search request information and to one or more search criteria to identify the one or more information items, the search criteria including that the identified information items each have an attachment to each topic in the accepted search request information.
20 . A carrier medium as recited in claim 19 , wherein the method further comprises:
displaying the results of the searching to the first searcher.
21 . A computer readable carrier medium carrying a data structure comprising:
a topic data structure configured to store a set of topics, the topic data structure having an element for each topic, the topic data structure further configured such that no topic hierarchy need exist for the set of topics;
a user data structure configured to store a set of users, a user being a searcher who has provided registration information, the user data structure having an element for each topic, each user data structure element including the registration information for the user;
an attachment data structure configured to store a set of attachments, each attachment being between one of the topics of the set of topics and an information item of a set of information items, the attachment data structure having an element for each attachment; and
the data structure configured for use by a computer implemented method of directing a search aided by the set of topics, the topics not necessarily having a hierarchy, the method comprising:
identifying for a first searcher one or more information items of the plurality of information items according to search request information provided by the first searcher, the search request information including at least one of the group consisting of a search phrase and a subset of one or more search topics of the set of topics;
determining one or more suggested topics from the set of topics, the suggested topics being determined according to the attachments of the suggested topics to the one or more identified information items; and
providing search results to the first searcher, the search results including at least some of the identified information items and at least one of the suggested topics,
wherein the suggested topics include one or more refinement topics determined from the set of topics according to a refinement topic criterion using a refinement selection method, and
such that the first searcher selecting one of the suggested topics generates a new set of suggested topics, such that a hierarchy of topics is formed on the fly for a particular search for the first searcher.
22 . A carrier medium as recited in claim 21 , wherein the data structure is in the form of a set of tables of a relational database, and wherein each of the topic data structure, the user data structure, and attachment data structure is in the form of a table of the relational database.