No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
The Application and Research of Classification of POCP SMS based on Text Mining
Abstract
People's opinion collection and processing system (POCP) is the electronic petition system with the function of website, hot line, E-mail and short message service (SMS), which is developed for Fuzhou Municipal Committee in JiangXi province. With the increasing demands of government service quality, the number of opinions brought by the masses is also on the increase. Seeking effective means and classifying these opinions to carry out contrapuntally become a gradually urgent problem. Based on the characteristic that SMS can change into text, this paper presents a thought that text classification algorithm can be used in the technology of SMS processing. SMS will be classified to different fields by preprocessing, feature section and classifier. At last, the reliable bases to government decision-making will be provided contrapuntally by analyzing the field classification of opinions.
... In [7], a probabilistic knowledge base is employed to conceptualize short text with applications to clustering Twitter messages. Much attention has also been paid to the mining of opinions [8] ...
Short messages are an important part of social media network. Besides traditional communications, short messages can also be used for education, entertainment and even business. Nowadays, much work has been devoted to mining useful information and knowledge from short messages via classification and clustering approaches. However, in many situations, it is vital to identify some influential users whose messages will result in a large amount of forwarding, these users are called seed users. In this paper, studies on the forwarding behavior of mobile phone users are carried out first, then, a directed graph-based short message forwarding model is setup, together with a "Reverse PageRank" scoring policy. Finally, the forwarding model and scoring policy have been applied to a real short message record dataset from some telecom operator. By some exploratory data analysis of the dataset, parameters of the forwarding model are determined. Experimental results show that the model and the scoring policy are effective in finding out seed users.
In view of shortcomings in the traditional digital campus, mobile digital campus has become the object of research. Using a mobile phone message construct SMS Platform for Campus, with mobile phone carriers, the existing system of backend management transform for the front-end service to provide user. This paper uses the mobile OpenMAS system, from the demand analysis, system design, main functions, and introduces how to use SMS platform to provide better information and services for mobile digital campus. Due to the adoption of a mature SMS middleware, it simplifies the difficulty of SMS application development.
Extraction of evaluation object has become a hot issue in the field of natural language processing. We combine semi-supervised and unsupervised Bootstrapping to extract Uyghur evaluation object in Web documents. In the pattern selection, we extract the information of words and their POS from 3 to 5 windows including seed word to generate candidate patterns. Moreover, we come up with redundant pattern mechanism, and deals with candidate pattern combined with pattern scoring algorithm. In pattern matching facet, Fuzzy Matching and Multiple Matching approaches are used. In a word, the correct rate of this method in Uyghur automatically extraction experiment is 81.03%, the recall rate is 83.64%, and the F-measure is 82.32%. Experimental results show that the effectiveness of automatic extraction experiments of Uyghur evaluation object.
Recently, with an increasing number of major events spreading all over the Internet, the research for online hot topics detection system has been paid more and more attention. In this paper, we proposed an unsupervised and efficient hot topics detection approach, which is based on an improved ant colony text clustering (IACTC) algorithm. In view of the deficiencies of the basic ant colony text clustering (BACTC) method, the inverse tangent function is introduced into the proposed algorithm as a novel probability function, which could be flexibly adjusted. Meanwhile, in order to reduce ants' blinding movement, we add the memory organ to each ant, and we develop the concept of adaptive moving range. Additionally, we design a practical system framework with technical details to solve the problems involving in the real-time hot topics detection system (e.g., information collecting, web preprocessing and text processing). This system is also used to evaluate our proposed method.The experimental results show that our approach has numerous advantages and achieves satisfying results.
This paper studies on real simulation of the e-mail worm propagation behavior. E-mail worm is one of the most important reasons that lead to the e-mail spam, and dramatically decrease the computer network of its efficiency. Thus the system resources have been seriously damaged. In order to simulate the email worm in a better way, we have proposed an e-mail worm propagation model based on the combination of the user behaviors and the e-mail network topology, with which we can describe the email users of their behaviors through their probability of opening and checking their e-mails. This method has accurately simulated the spread of the worms in the network topology. The results of the simulation experiment have shown that this method can detect the characteristics of the worm propagation more effectively. And then we compared and analyzed the effect of the worm propagation, which is of significance in further study of the worm defense.
Service-oriented applications are increasingly requested to be adaptive and reliable, to face harsh market competitiveness and volatility. Towards achieving that, service developers must foster current adhoc and rigid available technology, e.g. WSDL, BPEL and WS-CDL, with early formal specification and validation and equip them with adaptability as steering forces. In paper puts first milestones towards such challenging foundations. More precisely, we first propose ECA-driven architectural interconnections to handle service composition logic. Then, we leverage this composition logic with aspect-oriented techniques towards coping with dynamic adaptability. As a core further step, we semantically govern this aspectual service composition modeling with a tailored rewriting logic theory. More precisely, we extend to Maude language to faithfully specify and validate this model.
Personal communication service (PCS) networks offer mobile users
diverse telecommunication applications, such as voice, data, and image,
with different bandwidth and quality-of-service (QoS) requirements. This
paper proposes an analytical model to investigate the performance of an
integrated voice/data mobile network with finite data buffer in terms of
voice-call blocking probability, data loss probability, and mean data
delay. The model is based on the movable-boundary scheme that
dynamically adjusts the number of channels for voice and data traffic.
With the movable-boundary scheme, the bandwidth can be utilized
efficiently while satisfying the QoS requirements for voice and data
traffic. Using our model, the impact of hot-spot traffic in the
heterogeneous PCS networks, in which the parameters (e.g., number of
channels, voice, and data arrival rates) of cells can be varied, can be
effectively analyzed. In addition, an iterative algorithm based on our
model is proposed to determine the handoff traffic, which computes the
system performance in polynomial-bounded time. The analytical model is
validated by simulation
From a resource-based point of view, firm’s technological capabilities can be used as underlying sources for identifying new businesses. However, current methods are insufficient to systematically and clearly support firms in finding new business areas based on their technological strength. This research proposes a systematic approach to identify new business areas grounded on the relative technological strength of firms. Patent information is useful as a measure of firms’ technological resources and data envelopment analysis (DEA) is beneficial to obtain the weighted value of patents according to their quality. With this weighted quality of patents, a firm can evaluate their relative technological strength at the industry and product level according to potential business areas. To compute technological strength by products, this research applies text mining method to patent documents, a method which a researcher discovers knowledge with unstructured data with. This paper shows the usefulness of the newly proposed framework with a case study.