Project Risk Management part 3
This is a solution of Project Risk Management part 3 in which we discuss Developing business strategy can help your company cope with aging systems and limited resources that can lead to fragmented IT
Project Risk Management part 3
SYSTEM ARCHITECTURE
This section briefly describes the system architecture for the EWAS for monitoring news articles. News covers everything about anything in world. News update readers can read about the direction in which his/her interested entity in the world is moving. The interested entity can be the whole country, a bureaucratic session, a terrorist organization or anything. Services which analyze and read thousands of news items from heterogeneous sources and provide consolidated channels to access the news have become available. The biggest advantage of using such a service is that the duplication issues, topic identification issues, and other linguistic problems are being solved during the consolidation process, thus providing filtered information for analysis. Europe Media Monitor (EMM) is one of these services and is being used in this project [1]. EMM provides an RSS feed of news items which contain all of the consolidated news items. RSS adds some structure to the news by specifying its publication info but still the facts and nuggets lie buried in the text which actually forms the news.
Text Classification Process and Analysis of Algorithms
The project described in this paper used data from construction interorganizational information systems to evaluate text classification algorithms and to guide the development of the prototype of a construction document classification system. Bricsnet Inc., a company that has developed commercial Internet-enabled project management and collaboration services specifically for construction, has provided access to data from current and past projects that used its services. Bricsnet’s servers are used to host construction project extranets, creating a communication environment for project participants and providing a common repository for project documents. One construction project database was selected as a case study. The database was used by 16 project team organizations and contained around 4,000 document files ~1.5 GB!. Several types of construction documents were available in this database, including specifications, meeting minutes, requests for information, architect’s supplemental instructions, change orders, and field reports, among others. Meeting minutes were selected for this evaluation, which store information about weekly progress meetings among project participants. Each topic discussed during a meeting is recorded in separate items, which are grouped by topics according to specific project divisions. The divisions for this particular project were
• A–GENERAL;
• B–SCHEDULE;
• C–DEMOLITION-CIVIL;
• D–LANDSCAPE-SITE;
• E–STRUCTURES;
• F–BUILDING–SKIN;
• G–ROOFING-WATERPROOF;
• I–INTERIOR FINISHES;
• M–CONVEYANCE;
• N–PLUMBING;
• O–FIRE PROTECTION;
• P–HVAC; and
• Q–ELECTRICAL
Originally there were 92 meeting minutes. Each item for all of these meeting minutes was automatically extracted from the original document and stored in separate document files. A total of 845 documents were then used in the document classification process, which followed the steps described and detailed next, which range from data selection to creation of classification models, followed by discussion of the results. Data Selection In the first step, classes were defined as the project divisions presented previously. This step also involved selection of the documents used to create the classification model. In this step, the training positive, training negative, testing positive, and testing negative documents for each class were selected. These documents could be stored in central databases, such as the project extranets, or in distributed databases. Currently, only the case in which the data are stored in a central location is being explored. Data Preparation Text-based information is usually stored using formats such as word processor, spreadsheet, e-mail, HTML, XML, Postscript ~PS!, or Adobe Acrobat Portable Document Format ~PDF! files. In order to apply the preprocessing and classification algorithms, these files must be converted to text file format. This procedure used file converter computer systems to create a text version copy of each document, while keeping the original documents in their native formats and locations. The text versions were then used in the remaining steps of the classification process.
• A–GENERAL;
• B–SCHEDULE;
• C–DEMOLITION-CIVIL;
• D–LANDSCAPE-SITE;
• E–STRUCTURES;
• F–BUILDING–SKIN;
• G–ROOFING-WATERPROOF;
• I–INTERIOR FINISHES;
• M–CONVEYANCE;
• N–PLUMBING;
• O–FIRE PROTECTION;
• P–HVAC; and
• Q–ELECTRICAL
Originally there were 92 meeting minutes. Each item for all of these meeting minutes was automatically extracted from the original document and stored in separate document files. A total of 845 documents were then used in the document classification process, which followed the steps described and detailed next, which range from data selection to creation of classification models, followed by discussion of the results. Data Selection In the first step, classes were defined as the project divisions presented previously. This step also involved selection of the documents used to create the classification model. In this step, the training positive, training negative, testing positive, and testing negative documents for each class were selected. These documents could be stored in central databases, such as the project extranets, or in distributed databases. Currently, only the case in which the data are stored in a central location is being explored. Data Preparation Text-based information is usually stored using formats such as word processor, spreadsheet, e-mail, HTML, XML, Postscript ~PS!, or Adobe Acrobat Portable Document Format ~PDF! files. In order to apply the preprocessing and classification algorithms, these files must be converted to text file format. This procedure used file converter computer systems to create a text version copy of each document, while keeping the original documents in their native formats and locations. The text versions were then used in the remaining steps of the classification process.
Cheap Assignment Help UK provide assignment writing service based on case study requirements in Project Risk Management are providing most flexible online assignment writing help, so book your Assignment with us, order now
Comments
Post a Comment