A System for Dialed Number Prediction on Mobile Phones Based on Call History S.R. Subramanya School of Engineering and Computing National University San Diego, CA, USA Abstract— There are several instances where mobile phone users would not have stored all the numbers in their phone book (or PIM – personal information management). Even if a phone number is stored in the PIM, users would want to make a phone call using the keypad, without scrolling down the names in the PIM. The system proposed in this paper provides the users with the convenience of ‘number completion’ based on the first few digits keyed in. The system uses a number completion heuristic which makes use of the call history and the context of the user, and makes a prediction and suggests a few numbers. The short list of phone numbers suggested by the system is expected to contain the user–intended phone number with high probability. This saves the users from entering the whole numbers or searching through the call log (especially in a hurry), and avoid common mistakes (such as flipping of digits). It facilitates fast and correct dialing even in time constrained situations, without distracting the user for a long time. Keywords—mobile phone dialing; call history; number prediction; mobile phone user experience I. INTRODUCTION We have been witnessing the profound growths of the Internet, mobile technologies, and mobile Apps in recent years. The mobile phones, especially, have become ubiquitous rapidly, and are making profound impacts on almost all facets of our lives. The penetration of the mobile phone technologies has been so much in so little time, there has not been any other technology in human history like that till now. While it took landline telephones about 45 years to get from 5 percent to 50 percent penetration among U.S. households, the mobile phones (considering both feature phones and smartphones) took around seven years to reach a similar proportion of consumers. The penetration of smart phones have been even more rapid – having gone from 5 percent to 40 percent in about four years, [1]. As concrete examples, in 1982, there were 4.6 billion people in the world, and not a single mobile–phone subscriber. In 2015, there were about 7.35 Billion people in the world [2], and about 7.33 Billion mobile connections worldwide [3]. There are now at least 1.4 billion PCs in use worldwide. Mobile phones, on the other hand, are already selling more than 1.4 billion units every single year [4]. Since the middle of 2008 when Apple’s App store and Google Play were launched, as of June 2016, there were about 2 Million iOS Apps available in Apple’s App store, and about 2.2 Million Android Apps available in Google Play. The number of cumulative downloads have been 140 Billion for iOS Apps (as of September 2016), and 65 Billion for Android (as of May 2016). The number of active Apps by 2020 is expected to be about 5 Million [3]. The mobile phones have transformed rather quickly from simple voice–communication devices to multi–faceted devices providing voice, text, and video messaging, having built–in cameras for picture and video capture and playback, audio recording and playback, functionalities of clocks, compass, calendar, e–mail, Web browsing, etc. Although the mobile phones (smart phones) have tremendous processing power (many models having quad core processors) and memory (several GB), they have a few inherent limitations (compared to desktops and laptops) such as small screen, short sessions, single window visible at one time, limited battery power, and variable connectivity. These constraints will necessarily be present since the sizes of mobile phones cannot become bigger and/or heavier. These constraints result in mobile users (a) incurring a higher interaction cost in order to access the same amount of information; (b) relying on their short–term memory to refer to information that is not visible on the screen. These are also referred to as ‘physical’ and ‘cognitive’ overloads. The mobile users are usually in time–constrained, space–constrained, and attention–constrained situations. It is extremely important for the mobile phone designs to strive for reducing the above overloads by avoiding cramming too many features into them. II. BACKGROUND (USER INTERFACE AND EXPERIENCE) An observation of the mobile phone industry shows that the components of most of the mainstream phones (the chipset, the outer body, camera, screen, etc.) are very similar and commoditized, and the distinctions and variations between the phones in the above aspect are finer and subtle. However, the user interfaces and the features are rather distinct, and provide varying levels of user experience. The constraints of mobile devices (predominantly smaller screen size) pose several challenges in the design of user interfaces. User interfaces for mobile devices play a crucial role in reducing the above mentioned overloads and facilitating a pleasant and rich user experience. This is crucial in sustaining the growths in the consumption of content, and in the subsequent increases in consumption of mobile content and use of associated services. Several aspects of user interfaces for mobile content are presented in [5]. In general, the user interface should be intuitive and easy-to-use. The importance