Department of Computer Science

Permanent URI for this collection

https://escholar.umt.edu.pk/handle/123456789/208

Browse

Browsing Department of Computer Science by Author "Abbas Chishty Ansar"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Incidents Management System (IMS) – A software for management of Human Loss News Reports
    (University of Management and Technology Lahore, 2017) Abbas Chishty Ansar
    In today's world, with the exponential rise of human population and due to emergence of modern technology, every society is facing such challenges that they have never faced before during entire course of history. One of the main challenges among many is to secure and protect lives of subjects from any kind of emergency deaths and injuries. Human lives are lost or damaged due to accidents, disasters and various criminal activities and during law and enforcement agencies operations at regular pace in every society. This loss is priceless and very sensitive matter for every society. One way of protecting people from such situations is to take preemptive actions so that such incidents can be made little to happen. For this purpose every nation sets up departments those plan different strategies to prevent from such situations. These departments and their personals require different tools and technologies to assist them in their planning's and completion of relevant operational tasks. We are living in the age of modern science and a unique miracle of modern age science is information technology. We are in the realm of information technology. Back bone of informational applications is data. Different kind of data is generated every day in every society. Media is considered fourth pillar of society and its role in different walks of daily life is above doubts and non-debatable. In every society lot of accidents, crimes and other incidents happen that involve loss of human lives. These incidents do not go unnoticed. These are reported by different segments of society and regular media is one of them besides many social media platforms. Newspaper agencies are one of regular media outlets. There reports are reliable and trust worthy. Newspaper websites are modern form of newspapers. These are easily accessible and updated at regular intervals as compared to old classical styled printed newspapers. Archives of these news websites are easily and widely accessible at zero cost. Pakistanis a nation of worth 200 million inhabitants in the north western region of South Asia. It is passing through a very crucial period of its history. Besides regular incidents of human loss like every other nation, it is also bleeding by the hands of many terrorist organizations operating in Pakistanand many other neighboring countries of the region. It is been more than a decade that Pakistani nation is fighting against these terrorists. Almost every day terrorists target different segments of this nation. It is need of the day to curb such activities, data about such incidents must be captured, collected in an organized way, analyzed in a systematic way to find different useful information, share this information among different departments dealing with these events in a joint and organized manner. So that relevant authorities can deal with these situations proactively and effectively. In this thesis, we proposed a state of the art framework known as Incidents Management System or IMS, to tackle the challenges described above. The proposed framework can be helpful in addressing and avoiding human loss for authorities, as this system can help them incollecting, processing and analyzing the real-time data about human loss reports. The historical data of all such reports would also be stored in an efficient repository for analysis. We developed a prototype of many of the key components of the proposed framework crawler, classifier and reporter. We evaluated various classifiers for this problem against a middle sized data of 1600 plus reports. A number of performance metrics are used to measure the performance of these classifiers such as MNB, SVM, KNN and RF.The prototype of this system demonstrated the successful collection and analysis.