國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Chapter 5 Result and discussion
5.1 Cross analysis of LBS in different data features
From previous chapter 4, this present research already introduced and discussed the different 8 cases (within 13 applications). At literature review, we presented the 6 items of data features of LBS with big data. The items of feature are Routing path, Non-structure, Imprecision and Varying Precision, Digital data, Real Time, and Social network. The detail of each item are described inside the content of chapter.
This research found all cases are digitalized data and real time data. May not include routing path, but contain lots of location data. Precision situational are most factors. Non-structured are data of various, defined, pre-defined, un-defined. Some are collected from social network, some are not.
The relation between 6 items of factor with each of 8 cases are not complete synchronized to each other’s. Some cases are fully synchronized with 6 factors. Some cases are partially synchronized. For example, case 1.Taxi Fleet location and dispatch system and case 3. Chinese New Year Mass Migration with Baidu Heat Map are fully synchronized with all 6 factors. Case 6. Vehicle monitoring and tracking system and case 8. Navigation Services are strongly synchronized with partial of 5 factors. Case 2. Real-Time Traffic Information Service and case 4. Smart Logistic Network and case 5. Car pool network system and case 7. Public bicycle sharing system are synchronized with major factors.
The relation of case and features are judged by collected literature and case materials.
The result is depend on what features of case have. Item by item to review each factor, whether they are relative. For example, case 1 and case 3 are fully synchronized with 6 items of feature.
Because of they collected the data of routing path, the data are non-structure, the data accuracy are imprecision and variety, digital data was generated, data transfer in real time, and users are group in social network. The other cases are using same methodology to check each item of features whether they are relative.
Below table showed the results of data features with 8 studied cases. The block of table is marked with “X”, which means they are relative to each other.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 5-1 Data features of studied cases
1. Taxi Fleet location and dispatch system 2. Real-Time Traffic Information Service 3. ChineseNew Year Mass Migration with Baidu Heat Map 4. Smart Logistic Network 5. Car pool network system 6. Vehicle monitoring and tracking system 7. Public bicycle sharing system 8. Navigation Services 1. Data Features:
Routing path
X X X X X X
Non-structure
X X
Imprecision and Varying
Precision X X X X X
Digital data
X X X X X X X X
Real Time
X X X X X X X X
Social network
X X X X X X
5.2 Related technologies with Data Mining
From previous chapter 4, this present research already introduced and discussed the different 8 cases (within 13 applications). At literature review, we presented the 6 items of related technology of LBS with big data. The items of feature are Schema-less databases, Cloud computing, MapReduce, Hadoop, Hive and PIG, and Storage Technologies. The detail of each item are described inside the content of chapter.
This research found all cases must involve cloud computing and storage technology.
Other technologies are applied depended on structure of data and computational needs.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
The relation between 6 items of technology with each of 8 cases are not complete synchronized to each other’s. Some cases are strongly synchronized with most of technologies.
Some cases are partially synchronized. For example, case 3. Chinese New Year Mass Migration with Baidu Heat Map and case 4. Smart Logistic Network are strongly synchronized with 6 technologies. Case 1. Taxi Fleet location and dispatch system, case 2. Real-Time Traffic Information Service, case 5. Car pool network system, case 6. Vehicle monitoring and tracking system, case 7. Public bicycle sharing system and case 8. Navigation Services are strongly synchronized with major factors.
Below table showed the results of technology involved with 8 studied cases. The block of table is marked with “X”, which means they are relative to each other.
Table 5-2 Technology involved of studied cases
1. Taxi Fleet location and dispatch system 2. Real-Time Traffic Information Service 3. ChineseNew Year Mass Migration with Baidu Heat Map 4. Smart Logistic Network 5. Car pool network system 6. Vehicle monitoring and tracking system 7. Public bicycle sharing system 8. Navigation Services 2. Technology involved
Schema-less databases
X X X
Cloud computing
X X X X X X X X
Storage Technologies
X X X X X X X X
Hadoop
X X
Hive and PIG
X X
MapReduce
X X
In other point of view, we found the 6 technologies also can classified into two portions of technology. The front of 3 items, Schema-less databases, Cloud computing, and Storage
‧
Technologies, which provide operational capabilities for data structure, real-time computing, interactive data stored, where data is primarily captured and transit. The rest of 3 items which provide analytical capabilities for deep and native support that may touch data transit node, querying data in database, and mapreduce to simplify data of the collected data. These technologies are complementary and frequently deployed together.
Operational systems, such as the NoSQL databases, database products were built for speed and capacity. The database is highly scalable and capable of being used as a standalone database for queries. The service can be implemented very quickly and can interactively query huge amounts of data. It is in use on data tables that have hundreds of billions of rows.
Analytical systems, software is interactive. Users access and analyze very large amounts of data interactively, they were manipulating data in a spread sheet or using a statistical analysis tool. Systems tend to operate over many servers. Operating in clustering, managing hundreds of terabytes of data across billions of records. (MongoDB, 2014)
5.3 Efforts in big data management
From previous chapter 4, this present research already introduced and discussed the different 8 cases (within 13 applications). At literature review, we presented the 3 items of management efforts of LBS with big data. The items of effort are Database management, Clustering Data, and Recommendation system. The detail of each item are described inside the content of chapter.
This research found may not go through specific database management and clustering technique. But, recommendation process is a must.
The relation between 3 items of efforts with each of 8 cases are not complete synchronized to each other’s. Some cases are fully synchronized with all of efforts. Some cases are partially synchronized. For example, case 1. Taxi Fleet location and dispatch system, and case 4. Smart Logistic Network are fully synchronized with 3 efforts. Case 2. Real-Time Traffic Information Service, case 3. Chinese New Year Mass Migration with Baidu Heat Map, case 5. Car pool network system, case 6. Vehicle monitoring and tracking system, case 7. Public bicycle sharing system and case 8. Navigation Services are strongly synchronized with major factors.
Below table showed the results of management efforts with 8 studied cases. The block of table is marked with “X”, which means they are relative to each other.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 5-3 Management effort of studied cases
1. Taxi Fleet location and dispatch system 2. Real-Time Traffic Information Service 3. ChineseNew Year Mass Migration with Baidu Heat Map 4. Smart Logistic Network 5. Car pool network system 6. Vehicle monitoring and tracking system 7. Public bicycle sharing system 8. Navigation Services 3. Management effort
Database management
X X X X X X
Clustering Data
X X X
Recommendation system
X X X X X X X X
In every industry, top managers wonder whether they are getting full value from the massive amounts of data of location-based they already have within their organizations. New applications of location-based service are generating more data than ever before, and also better technologies of big data are solving the issues to improving management efficient. Many businesses are still looking for better ways to obtain value from their collected data then resulting in the marketplace (LaValle et al., 2011).
Leveraging that data to drive smarter decisions requires new thinking. The companies adopt big data as a strategy to improve efficient, Decision platform is becoming a key factor of business operation.
5.4 Develop value from big data
Based on previous section, which identified the results of asked question performed by the studied cases. This section will make a conclusion of created value from big data by using business model framework proposed by Alex Osterwalder, showed on the chapter 2-3. The framework was separated into 4 major blocks, which are 1) the offering. 2) Infrastructure. 3)
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Customers. 4) Finances. The dimensions of the LBS business model framework were derived from a systematic review of studied case, measured by the number of citations, as shown in Chapter 4.
This research found mainly in customer and infrastructure components. “Customer” can experience significant value from the LBS big data and whole business “Infrastructure” can also receive big improvement in the back of office process.
The relation between 4 blocks of business model with each of studied cases are not complete synchronized to each other’s. Some cases are fully synchronized with all of blocks.
Some cases are partially synchronized. For example, case 1. Taxi Fleet location and dispatch system, case 4. Smart Logistic Network, and case 6. Vehicle monitoring and tracking system, are fully synchronized with 4 blocks. Case 2. Real-Time Traffic Information Service, case 3.
Chinese New Year Mass Migration with Baidu Heat Map, case 5. Car pool network system, case 7. Public bicycle sharing system and case 8. Navigation Services are strongly synchronized with major blocks.
Below table showed the results of Business Model framework with 8 studied cases.
Table 5-4 Develop value of studied cases
1. Taxi Fleet location and dispatch system 2. Real-Time Traffic Information Service 3. ChineseNew Year Mass Migration with Baidu Heat Map 4. Smart Logistic Network 5. Car pool network system 6. Vehicle monitoring and tracking system 7. Public bicycle sharing system 8. Navigation Services Business Model review
1) The offering
X X X X
2) Infrastructure
X X X X X X X X
3) Customers
X X X X X X X X
4) Finances
X X X
‧
Learned from chapter 5, the results showed the 8 studied cases related to each research objectives. We have answered each of the questions proposed on chapter 1.3. It helps business management team, how to leverage the massive data created from location-based service. This study provided 8 different case for enterprise to refer to their own operation.
The materials of the LBS literature were derived from a systematic review, measured by the number of citations, as shown in Chapter 2. The whole research followed the research method in the chapter 3. Including content analysis, data collection, and data analysis.
In order to understand the management of LBS big data, this study collected 8 different type of LBS cases in chapter 4. Each of the case was described about the major applications of current marketplace. The content focus on the answering research questions, including of the features, collection and analysis methods of the big data on LBS and examine the management efforts and benefits of the use of big data results from LBS.
A very large number of data has been generated off those LBS. The LBS data sets are usually generated in real time, in large volumes and in unstructured forms. This study already described how the data features are, what technology was used, and what efforts done by management.
This research found all cases are digitalized data and real time data. May not include routing path, but contain lots of location data. Precision situational are most factors. Non-structured are data of various, defined, pre-defined, un-defined. Some are collected from social network, some are not. All cases must involve cloud computing and storage technology. Other technologies are applied depended on structure of data and computational needs. All cases may not go through specific database management and clustering technique. But, recommendation process is a must. The value was developed mainly in customer and infrastructure components.
“Customer” can experience significant value from the LBS big data and whole business
“Infrastructure” can also receive big improvement in the back of office process. The tables were collected and put on the content of chapter 5.
6.2 Contribution and management implication
The main contribution of present research is that it will help creating an advantage and