GB2536504A

GB2536504A - System for choosing data

Info

Publication number: GB2536504A
Application number: GB1504802.8A
Authority: GB
Inventors: Henry Matthew Stuart-Smith James
Original assignee: Individual
Current assignee: Individual
Priority date: 2015-03-20
Filing date: 2015-03-20
Publication date: 2016-09-21
Also published as: GB201504802D0

Abstract

A system for choosing data to be displayed on a first device includes a first database comprising an autocomplete dictionary, a second database listing data to be displayed on the device and selecting means for selecting from the said second database data to be displayed at any given time, characterised in that the first database includes a vocabulary of words commonly used by the user of the device and ranked or weighted by their frequency of use and that the said selecting means is configured to access the said words in the first database together with their rank or weighting and select from the second database data to be displayed based partly or wholly on the said words and their rank or weighting.

Description

SYSTEM FOR CHOOSING DATA

The current invention relates to methods of choosing data to be displayed on the screen of a computer, tablet, mobile device or other device having a screen.

Current screen based devices such as computers, televisions, tablets, mobile phones and games consoles commonly divide the "real estate" of the screen in some modes, often dedicating some of the screen availability to advertising. There is known a multitude of ways of doing this, including a bar down one side of the screen, pop-ups that appear in the middle of the screen and full-screen displays that appear at the end of a game, for example. Depending on the device, this may be controlled at any one time by the operating system, an application, including email or social media, or a browser.

Whatever the method of display, there remains a question of how to choose advertisements to display which will be appropriate for the user of the device. Various methods are already known. For example a games app is likely to advertise other games that would appeal to a player. It is also well known to look at browser cookies to see what websites the user has been visiting and to display advertisements relating to those websites. A problem arises with such a method. Because it uses data from sites already visited, it often results in advertisements for products and services that the user has already bought. For example, if the user buys a watch, the method described will cause advertisements for watches to be shown, even though the user has already satisfied his/her current need.

Screen based devices also commonly include some sort of data input method. On more sophisticated devices this will allow the input of alphanumeric characters although less sophisticated devices may only permit access to menus. In the case of devices allowing input of alphanumeric characters the input method will usually be a keyboard, either of the full alphanumeric or "QWERTY" type, or of another type such as a telephone keypad.

In either case there is commonly an autocorrect or autocomplete function incorporated in the device. This aims to reduce the effort of typing by completing -in the case of autocomplete -or correcting -in the case of autocorrect words as they are typed.

According to the present invention there is provided a system for choosing data to be displayed on a first device the system including a first database comprising a dictionary, a second database listing data to be displayed on the device and selecting means for selecting from the said second database data to be displayed at any given time, characterised in that the first database includes a vocabulary of words commonly used by the user of the device and ranked or weighted by their frequency of use across more than one application on the device and that the said selecting means is configured to access the said words in the first database together with their rank or weighting and select from the second database data to be displayed based partly or wholly on the said words and their rank or weighting.

Preferably the said first database is the default dictionary for the device.

Preferably the said first database is used as a dictionary by several applications on the device.

Preferably the said selecting means is configured to select data for display that includes one or more of the said words.

Preferably the said selecting means is configurable by the user to include or exclude specific words chosen by the user.

Preferably the system includes a correlating database which correlates websites visited by multiple devices with the rank or weighting of words used on those devices and said selecting means is configured to select data to be displayed by comparing the rank or weighting of words on the said first device with the rank or weighting of words in the correlating database.

Preferably the said correlating database is the said second database. Preferably the said first database is stored on the device.

Preferably the said selecting means is configured to select data by selecting a first group of data based on the rank or weighting of the said words and then making a further selection within that subgroup based on a second method.

Preferably the said second method is a sequential selection of data within the said subgroup.

Preferably the said second method is a random or quasi random selection of data within the said subgroup.

Preferably the selecting means stores some of the said words together with their weight or ranking in a further database remote from the said first database.

Preferably the selecting means stores a list of selected data to be displayed which list is updated from time to time.

The present invention will now be more fully described with reference to the accompanying drawings in which: Figure 1 shows a flow diagram of a system incorporating a first embodiment of the current invention Figure 2 shows the architecture of the system of Figure 1 Figure 3 shows a flow diagram of a system incorporating a second embodiment of the current invention and Figure 4 shows the architecture of the system of Figure 3 Figure 1 shows a flow diagram of a system incorporating a first embodiment of the current invention. In this embodiment the device has an autocomplete database installed on the device which acts as the default autocomplete database and is used by several applications across the device. The autocomplete goes through a first loop as the user types. Each time the user starts to type a word 401 the database is searched for words starting with the string typed or with common mis-typings that may result in the string typed 402. A selection of these words is displayed on the screen 403. The user can either select one of the words displayed or can continue to type the complete word that they meant to use 404. In either case, the database is updated 405 to reflect the choice of word, giving it a weighting based on its frequency of use by the user. This weighting is subsequently used to select words to be displayed when the same opening string is typed. For example, if the user types the word "band" very often, the system will display this word among the selections when B and A are typed, in preference to other words such as bad, banter or bandage, assuming these words are typed less commonly by the user. However if over time the user starts to type "banter" much more frequently, the respective weightings will adjust and "banter" will be displayed as a choice when B and A are typed, displacing another word from the list. Contextual weightings may also be applied in the words chosen to be displayed but these do not affect the frequency weighting of the words. Depending on the configuration of the autocomplete system, common words like "the" may or may not be weighted for frequency of use.

A second database, held in this embodiment off the device and accessed via a network, lists data to be displayed. The second database may also store the data to be displayed or it may provide links to the data which is stored in various places elsewhere. The data may be advertisements. Associated with each listed item of data are various keywords. These may be directly related to the data. For example a watch advertisement may have the word "watch" associated with it, along with other words. The keywords may also be less directly related to the data. For example, the word "outdoor" may be linked with an advertisement for running shoes.

The system also includes a selecting means, held in this embodiment off the device and accessed via a network. The selecting means from time to time scans the autocomplete database 501 and notes words which are recorded as having a high frequency of use and which are not common words like "the". It then compares these words with the keywords in the second database 502. Where there is a match between the noted words and the keywords the selecting means chooses the associated data 503 for display on the screen of the device. Commonly there will be several matches, in which case the selecting means may use other selection techniques to reduce the number of data items to be displayed and commonly also display the various data items either sequentially or at random.

To avoid the need for constant access to the autocomplete database the selecting means can either hold a copy of the database locally and update it from time to time or hold a selection of words with or without their weighting, which is again updated from time to time. Conveniently the selecting means may also store on the device a list of data items to be displayed which is updated from time to time.

By virtue of the fact that the autocomplete feature is used over more than one application the data provided gives a more complete picture of the user's interests than is achieved by monitoring only web traffic, for example. Consequently the user is more likely to be provided with data that is of interest to them. To avoid undesired data relating to specific interests from being displayed a user-controlled word filter may be included.

Figure 2 shows the architecture of a system incorporating the embodiment shown in Figure 1. A user 1 interacts with the user device 2. The device 2 includes an operating system 3. Running within the operating system is a system application 4 forming part of the current invention. The system application retrieves words together with ranking or weighting data from an autocomplete database 5 and transmits this data via the internet 6 to a selecting means 7 which compares the word data with data held on a second database 8. The selecting means then returns data to the user device 2 via the internet where the system application 4 allows it to be displayed. An autocomplete application 9 exchanges data with several applications 20 on the device and updates the autocomplete database 5, based on the entries typed by the user within any of the applications 20. It will be appreciated that the invention applies equally to a similarly operating autocorrect or other dictionary database and application, as already stated.

Figure 3 shows a flow diagram of another embodiment of the current invention. In this embodiment, the autocorrect process is the same as the embodiment of Figures 1 and 2. The selecting means downloads either a copy of the autocorrect database or a selection of words with or without their weightings, as in the embodiment of Figures 1 and 2. The selecting also receives from the device details of websites visited by the user. These details are anonymised. The selecting means collects the same data sets from a large number of devices 601, all anonymised. It then correlates in a correlating database the autocorrect words with websites visited over the data sets to provide a table of websites associated with keywords 602. The table may or may not include a weighting based on the degree of correlation found.

The table is then used to populate the keyword associations in a database corresponding to the second database in the first embodiment 603. This database may be a separate database to the correlating database or the two may be the same database. The keyword population may be a direct mapping, whereby keywords are linked directly to the sites with which they have a high correlation with visits. For example, if there is a high correlation between the word "Hearst" and visits to the Hearst castle website, then the keyword "Hearst" could be linked to the Hearst castle website. Alternatively there could be an indirect mapping, where the correlation between the word "Hearst" and the Hearst castle website results in a link to a site providing comfortable walking shoes.

Figure 4 shows the architecture of the system of Figure 3. The architecture is similar to that of Figure 2 but includes also on the device a store of browsing data 10 which is transmitted to the selecting means 7 via the internet. The selecting means is also connected to and able to retrieve word and browsing data from multiple other devices used by other users. The data from these many devices is correlated and stored in a correlating database 11. The selecting means uses data from the correlating database 11, which may be part of the second database 8, to populate the keyword associations in the second database 8.

Other embodiments of the invention may be envisaged which incorporate other means of refining further the selection of data using known means. For example the use of geographical data could be used to refine the selection further.

Claims

CLAIMS1. A system for choosing data to be displayed on a first device the system including a first database comprising a dictionary, a second database listing data to be displayed on the device and selecting means for selecting from the said second database data to be displayed at any given time, characterised in that the first database includes a vocabulary of words commonly used by the user of the device and ranked or weighted by their frequency of use across more than one application on the device and that the said selecting means is configured to access the said words in the first database together with their rank or weighting and select from the second database data to be displayed based partly or wholly on the said words and their rank or weighting.
2. A system as claimed in claim 1, characterised in that the said first database is the default dictionary for the device.
3. A system as claimed in claim 1 or 2, characterised in that the said first database is used as a dictionary by several applications on the device.
4. A system as claimed in any of claims 1 to 3, characterised in that the said selecting means is configured to select data for display that includes one or more of the said words.
S. A system as claimed in any preceding claim, characterised in that the selecting means is configurable by the user to include or exclude specific words chosen by the user.
6. A system as claimed in any preceding claim, characterised in that the selecting means stores some of the said words together with their weight or ranking in a further database remote from the said first database.
7. A system as claimed in any preceding claim, characterised in that the selecting means stores a list of selected data to be displayed which list is updated from time to time.
8. A system as claimed in any preceding claim, characterised in that the system includes a correlating database which correlates websites visited by multiple devices with the rank or weighting of words used on those devices and said selecting means is configured to select data to be displayed by comparing the rank or weighting of words on the said first device with the rank or weighting of words in the correlating database.
9. A system as claimed in any preceding claim, characterised in that the said correlating database is the said second database.
10. A system as claimed in any preceding claim, characterised in that the said first database is stored on the device.
11. A system as claimed in any preceding claim, characterised in that the said selecting means is configured to select data by selecting a first group of data based on the rank or weighting of the said words and then making a further selection within that subgroup based on a second method.
12. A system as claimed in any preceding claim, characterised in that the said second method is a sequential selection of data within the said subgroup.
13. A system as claimed in any preceding claim, characterised in that the said second method is a random or quasi random selection of data within the said subgroup.
14. A system substantially as hereinbefore described with reference to the accompanying drawings.