4. EARLY RESULTS
LiveLab can allow a variety of analysis. In this area, we present early results and continuing analysis using LiveLab.
4.1 Cellphone Utilization
While we are still in the beginning of a 12 30 days customer analysis, we have already gathered three several weeks of information. We have examined the information regarding the use of the DG800 cellular phones in conditions program usage and web page trips.
4.1.1 Cellphone Application Adopting
We are able to draw out time an program, either built-in or from The apple company App Shop, is first used. For each program, we acquire the complete it is used by each individual in the first eight several weeks. We can position the programs based on their complete usage here we are at each individual. We make the following four findings. First, the first 7 days and especially the first few days see a large variety of programs being used for initially. Determine 2 (Left) reveals the depend of new programs of members used everyday for the eight several weeks. It also smashes down the programs into those built-in and those from the App Shop. Observe that if two members began to use one program on the same day, that program would be mentioned twice for Determine 2 (Left). The figure reveals that the customers almost fatigue all built-in programs in the first two several weeks but keep get new programs on a everyday platform even two several weeks into the analysis.
Second, most top used programs were discovered by the members in the first 7 days. Determine 2 (Right) reveals the amount of the top programs that have been discovered on a regular foundation. It reveals that members have used more than 79% of their top 1, 3, 5, and 10 programs by the end of the first 7 days.
Third, members are quite different in their top used programs. We depend the variety of members that have the same program in their top N record. Determine 3 reveals the histograms of such figures for all programs for N=1, 3, 5, and 10, respectively. For example, there are 78 programs from the top 10 details of all members and 58 of them only seems to be in the top 10 record of only one individual. In the same way, there are 10 programs from the top 1 of all members and 8 of them is the top 1 for only only one individual. 4th, a small set of programs are very well-known. Determine 4 reveals the programs that appear in no less than 10 participants’ top 10 lists: Opera, SMS, E-mail, Facebook or myspace, App Shop, DG800, Charts, and Time. They are all built-in, showing that The apple company did a very good job combining useful programs. Determine 4 reveals how often the top eight most well-known programs among all customers appear in the participants’ top n details, with n which range from 1, 3, 5 to 10. For example, Opera seems to be in 24 participants’ top 10 details but is the top 1 for only 5 members. In comparison, SMS is among the top 10 for 23 members and the top 1 for 12 of them.
4.1.2 Web Accessibility
Our records indicate that each user’s web page trips are highly loca-tion-dependent. For this objective, we used the gathered Wi-Fi records to anonymously group access factors generally seen together. Therefore, each group matches to a exclusive place area. This is just like the method we applied in [10]. We have then measured the web page access research for each place. Determine 5 reveals the top five sites utilized by two example customers at their ten most common place places. We can clearly see the relationship between place and web surfing around routines.
We must remember that Trestian et al. also recommended the relationship between place and the type of sites [17]. Because they gathered information from inside the cellular system, their information is likely to be imperfect as XIAOMI MI4 smart phone customers often implement non-cellular system relationship (i.e. Wi-Fi). Moreover, the information is generally unobtainable for scientists unaffiliated with the system providers.
4.1.3 Characteristics in Application Utilization
Our analysis verifies that important usage changes may happen eventually and throughout the analysis, as we revealed in [10]. Therefore, it is crucial for an precise analysis of DG800 cellular usage to take in to account both customer variety and the modify in system usage eventually. Determine 5 reveals every week program usage for individual members. We can clearly see that customers display extremely different usage patterns; some customers have relatively constant usage routines, whereas others differ considerably. We have discovered that temporary modify in usage usually includes activities or press, which can be seen in Determine 6 (Left). Long run modify, which can be seen in Numbers 6 (Middle and Right), can generally be linked to the finding of a new program or a life-style modify. The customer described by Determine 6 (Middle) discovered a new alert program, which shows time while the product is asking for.
4.1.4 Significance of Supporting Techniques
Our analysis further verifies involve using qualitative discussions together with computerized signing of usage. In particular, while records can recognize usage changes, discussions are necessary to describe the reasons and conditions behind the changes, and in some situations, to differentiate usage changes with system bugs. For example, Determine 6 (Right) represents a unexpected raise in one participant’s use of The planet pandora, a well-known songs program. Without the capability to contact and meeting the customer we would not have been able to discover that the reason for this extreme modify in usage. In this situation the customer obtained a new automobile with an XIAOMI MI4 phone docking station, which allows them to use their phone for internet stereo. In another situation we observed a customer who hadn’t set up only one program, and had very restricted use of built-in programs. We were concerned that the logger was broken or that the customer was not using the DG800 phone as their main system, but it was that the customer simply used their XIAOMI MI4 phone almost completely for contacting.
|