Introduction
The technological advances of the last few years have favoured the availability of increasingly more advanced mobile devices with high processing capabilities and enabled for access to digital contents of different kinds such as images, audio, video, etc.
Additionally, third generation networks provide users on the move with high bandwidth, of the order of megas, that they can use to download and interact in streaming mode with the different content types. Additionally, because there are more and more digital contents, these contents are now available also to users on the move.
One example of this is the .mobi domain that is dedicated to hosting websites that are accessible on the move. The .mobi initiative will reach its developmental zenith in 2007 with thousands of subdomains already bought up by major world companies. This will lead to a proliferation of mobile web contents to levels similar to what is now available over the Internet.
The following are the conclusions from the above claims:
- Mobile devices have become a thriving platform for the consumption of digital content
- A successful development of the contents market calls for users to get services and contents irrespective of time and place.
As happened with the Internet, providing advanced search technology on a set of mobile digital contents will improve and boost its use by users, achieving full development of the information society. However, searches on the move call for more sophisticated technologies than those used by conventional Internet browsers (Google, Yahoo, etc.), as the search results should be:
- Relevant for the time, place and preferences of the user making the search. For example, if the user is looking for bus stops, the search should return the bus stops closest to where the user is at the time
- Accessible from the user device. For example, if a user searches from his mobile device for a song, he or she should only be offered music in a format readable by his or her mobile device.
Additionally, the user should be able to launch searches without having to manually enter search terms, as mobile device usability is limited. The solution of this question will involve the use of technologies based on the semantic web concept.
The key goal of the MyMobileSearch project is to undertake R&D&I activities targeting the definition of mobile search technology capable of meeting the challenges set out above. Note that the project will not try to define a totally new or revolutionary search technology but will define new formalisms and techniques to integrate and extend today’s browsers (page-rank, semantic browsers, etc.) to bring their results and performance into line with the requirements set by searches on the move.
Project need
For all the reasons expressed above, there will be a major increase in business based on searches on the move. Searches on the move require technologies that go beyond the capabilities of conventional browsers (Google, Yahoo…) and are one of the opportunities for wealth creation in the coming years. The benefits of searches on the move are expected to exceed traditional Internet searches because:
- Mobile devices are always switched on, are always available and are always connected. End users can have access to information anywhere, at home, in the office, at a restaurant, or while they are on the move.
- It is very easy to contact people or businesses related to the contents output by the search.
- When they run searches on the move, users are actually where they are more likely to purchase a good or service.
- Although mobile searching bears similarities to some Internet search functions, users running mobile searches tend to be looking for an experience that offers direct responses and suitable results as opposed to the standard trial-and-error approach taken on the Internet, e.g. using Google.
- Mobile search applications should be able to assess the information about the user’s profile and the delivery context, making the search results as relevant as possible.
Mobile telephones are closely linked to users and representatives of the owner’s personality, indicating situational and behavioural patterns, as well as personal interests. The delivery context, which covers the user device, the mobile network capabilities, location, time of day, etc., is a key search parameter.
- The user should be able to launch searches without having to manually enter the search terms, as mobile device usability is limited.
On the other hand, the Web 2.0 phenomenon is driving more and more users to label the contents by means of folksonomies, which are collaboratively built taxonomies. This labelling is increasingly being used as an element to improve searches, of which gateways like Flicker, Youtube, Delicious or Bibsonomy are examples. Therefore, MyMobileSearch must take this new phenomenon into account and try to tailor it to the needs of searches on the move.
General objectives
The key objective of the MyMobileSearch project is to develop specifications, technologies and formalisms that improve and ease searches of contents on the move. To achieve this, it will be necessary to do R&D&I activities that permit the use of context-awareness as a key part of the search process. Context awareness includes aspects like user device, access network, location, time (time of the day and day of the week), situation (at work, shopping, etc.), as well as the user profile and user preferences and tastes. It will be essential to use Semantic Web based technologies to consider all these variables intelligently and take advantage of advanced reasoning capabilities.
Another MyMobileSearch objective is to integrate existing and mature search technologies into the spectrum of searches on the move. In no case does MyMobileSearch intend to develop search technology from scratch. Its aim to improve existing technologies to take into account context awareness aspects, improving their behaviour in environments on the move and, therefore, making the user experience more satisfactory. All this will stimulate the growth of information and contents consumption through mobile devices, developing the information society to its full potential.
Additionally, the soundness of the developed technology will be checked through experimentation consisting of the construction of a prototype and validation by real users. This will provide the feedback required to mark out MyMobileSearch’s evolutionary lines.
Finally, note that the project will be developed according to an open source software community scheme. This will provide dissemination, validation and contribution by other scientific and university members.
Specific objectives
The specific project objectives are detailed in the following:
- Specify the user and business requirements for searches on the move.
- Define the architecture of a mobile browser, defining the separation between the software elements that will be hosted by the user’s terminal and those that will be hosed at the server.
- Engineer ontologies to describe the delivery context and user profile that can be used during the search to filter results appropriately.
- Develop technology and formalisms for mobile browsers that provide search results that are:
- Relevant for the time, place, profile, social network and preferences of the user running the search.
- Accessible from the user device. For example, if a user searches a song from his or her mobile device, he or she should only be offered music in a format that is readable by his or her mobile device and that can be downloaded from the network to which he or she is connected at that time.
- Define and develop technologies and formalisms that ease user searches from mobile devices, providing maximum usability and at the same time increasing the likelihood of the search being successful.
- Study, specify and define the integration mechanisms between MyMobileSearch and
- Conventional browsers (based on the PageRank algorithm)
- Browsers based on formal semantics
- Browsers based on folksonomies
- Browsers based on microformats
- Georeferenced information browsers.
- Experiment using the development of a prototype (pilot test) of a mobile browser that will include:
- “MobileOK” robot concept test that is capable of indexing contents according to the downloading, interaction and visualization possibilities of the different device families
- Concept test of integration with a conventional browser
- Concept test of integration with a semantic browser
- Concept test of integration with a browser based on folksonomies
- Concept test of integration with advanced browser and user interface plugins that improve searching.
- Validate the solution with potential users of the technology.
Architecture
The architecture (to be explored and developed during the project) originally conceived for MyMobileSearch is reflected in the figure below.
Innovative aspects of the project
There are as yet no searching on the move solutions such as MyMobileSearch intends to build. Briefly, the most innovative aspects of the project are:
- Mobile searches through MyMobileSearch will be context and user profile aware, i.e. they will take into account aspects like the user’s device, access network, user preferences, tastes, current location, time of day, etc.
- Users will have the technology to support searches so that they do not have to enter terms by hand or have to apply tedious trial-and-error procedures.
- Like the Web 2.0 phenomenon, MyMobileSearch will take into account non-formal semantics provided by users (folksonomies).
- MyMobileSearch will be integrated with different browser types (conventional, semantic, folksonomy based).
