فایل ورد کامل مدل پیشنهادی برای داده کاوی توزیع شده تجارت الکترونیکی بر پایه SOAP و آنتولوژی

توجه : به همراه فایل word این محصول فایل پاورپوینت (PowerPoint) و اسلاید های آن به صورت هدیه ارائه خواهد شد

این مقاله، ترجمه شده یک مقاله مرجع و معتبر انگلیسی می باشد که به صورت بسیار عالی توسط متخصصین این رشته ترجمه شده است و به صورت فایل ورد (microsoft word) ارائه می گردد

متن داخلی مقاله بسیار عالی، پر محتوا و قابل درک می باشد و شما از استفاده ی آن بسیار لذت خواهید برد. ما عالی بودن این مقاله را تضمین می کنیم

فایل ورد این مقاله بسیار خوب تایپ شده و قابل کپی و ویرایش می باشد و تنظیمات آن نیز به صورت عالی انجام شده است؛ به همراه فایل ورد این مقاله یک فایل پاور پوینت نیز به شما ارئه خواهد شد که دارای یک قالب بسیار زیبا و تنظیمات نمایشی متعدد می باشد

توجه : در صورت مشاهده بهم ریختگی احتمالی در متون زیر ،دلیل ان کپی کردن این مطالب از داخل فایل می باشد و در فایل اصلی فایل ورد کامل مدل پیشنهادی برای داده کاوی توزیع شده تجارت الکترونیکی بر پایه SOAP و آنتولوژی،به هیچ وجه بهم ریختگی وجود ندارد

تعداد صفحات این فایل: ۹ صفحه

داده کاوی در محیط های تجارت الکترونیکی، یکی از برنامه های کاربردی داده کاوی توزیع شده (DDM) می باشد. برای طراحی یک مدل مناسب برای داده کاوی در محیط های تجارت الکترونیکی، می بایست مسائلی از قبیل داده های ناهمگن، حریم خصوصی، یادگیری داده ها، تعامل پذیری زیاد در ارتباط امن، و عملکرد را همواره مد نظر قرار دهید. مدل ECDDM پیشنهادی محتوی اجزای تخصیص داده شده برای حل این مسائل می باشد. در اینجا از بافرهای حافظه برای غلبه بر مسائل عملکرد و از آنتولوژیهای سلسله مراتبی برای اضافه نمودن معنا (سمنتی) به سطوح داده ها استفاده کرده ایم. در این مقاله، فرایندهای زمینه برای داده کاوی حفظ حریم خصوصی، ترکیب، و تجزیه پرس و جوهای داده کاوی پیشنهاد کرده و از تکنولوژی SOAP برای رسیدگی به برنامه های تجارت الکترونیکی ناهمگن، افزایش تعامل پذیری و پیام های ارتباطی امن استفاده کرده ایم. 

عنوان انگلیسی:Proposed Model for E-commerce Distributed Data Mining based on SOAP and Ontology~~en~~

Introduction Several important issues need to be addressed to design the ECDDM Model. The first one is the distributed nature and semantically disparate e-commerce customer data. Several approaches have been implemented for distributed data mining (DDM) [1]. One of these approaches is moving all distributed customer data into a central location and then performing mining tasks on the integrated data. Another approach is to perform mining tasks on each set of distributed customer data (for each location) to build local models. Then, local models will be moved to a central location and combined as a global model [2]. The final years of the past decade have seen the rapid development of learning classifiers from a semantically homogeneous relational database in the machine learning literature [3] [4]. In recent years, there has been an increasing interest to extend such approaches for learning classifiers from multiple semantically disparate, geographically distributed, relational data sources on the Semantic Web [5]. Privacy Preserving Data Mining (PPDM) is another important issue that needs to be considered during DDM model design. Most e-commerce providers may not be willing to share their data but they would like to take the benefits from DDM applications. According to [6], PDDM research is still in its infancy and there is no a practical system or development framework for PDDM. In recent years, there has been an increasing interest in data mining privacy methods [7]. Examples of these methods are: sanitation, data distortion, and Secure Multi-party Computation. Sanitation method aims to modify or remove sensitive data from data sources. Removing or modifying process may give a negative impact in the data mining results [8][9]. Distortion method provides privacy for e-commerce data by modifying the original data [2]. Another different method uses Secure Multi-party Computation (SMC) that uses cryptographic techniques to ensure almost optimal privacy [6]. The data learning process itself is a critical issue for this kind of model due to the relational nature of e-customer data. Relational nature of e-commerce customer data basically violates two assumptions made by traditional data mining techniques as stated by [4]: “The instances in relational data are not recorded in a homogeneous structure and are not independent and identical distributed”. In the ECDDM proposed model, we have used the approach that performs mining tasks for each distributed customer data to build local models and then combining them as a global model on the user side (central side). Statistical Relational learning is used as a classification process, and hierarchical ontologies are used to solve the problem of semantically heterogeneous data. There are privacy-based and performance-based components in both distributed customer data and user data. Simple Object Access Protocol (SOAP) is an XML-based protocol that will be used as a communication protocol to enable the user side to communicate with heterogeneous E-commerce applications. In section 2 the proposed model is discussed in more details. 2 ECDDM Proposed Model Structure 2.1 Overall Structure for the Proposed Model There are two sides for the proposed model. User side as central side and distributed competitor side. Various components in both sides work together as follows: Mining request is generated from user architectural components at the user side, decomposed to many SOAP requests for each competitor side and SOAP results will be composed to generate integrated results at the user side. Figure 1 shows the overall structure of this DDM model. SOAP adopts SSL encryption to encrypt the information, so it is secure to transfer information [10]. Figure 1: Overall structure for the proposed model 2.2 Distributed Competitor-Side Architectural Components As indicated in figure 2, there are three types of competitor side components. Physical component, coordination processes, and memory buffers. Physical component includes a local database or any other data source formats and algorithms library. Algorithms library contains all available algorithms for DDM and PPDM. Figure 2: Competitor-Side Architectural Components. Coordination processes are interfaces between physical components and memory buffers. Coordination processes includes SOAP Generator, ontology builder (OB), data miner (DM), data reader (DR) and Privacy Preserving Data Mining Process (PPDMP). The functions of these processes are as follows: SOAP Generator function: Receive “request SOAP message” and then convert it to competitor side environment. Convert mined result as “Response SOAP message” and send back to user side. Ontology builder (OB) function: It is responsible for Building hierarchical ontology in competitor side to be at the same abstraction level with user ontology. Hierarchical ontology will be constructed according to predefined mapping constraints between user side ontology and competitor side ontology.

$$en!!