Search Engines

Abstract

Nowadays, the volume of information on the internet is growing rapidly. Although this growth has many benefits, it is the reason for new challenges. The main challenge of the internet is how to retrieve the user’s requested information from a mass of information. This retrieval is done by search engines. Today, search engines are counted as one of the main components of the internet and we can hardly imagine the internet without them. To know how they work, first, we need to know their components and understand their tasks. In this workshop, we will briefly review how a search engine works and introduce algorithms used to retrieve information and page ranking which is done by search engines. After this workshop, you will have a general insight on search engines architectures and mainly used algorithms in search engines.

Headlines

Introduction to Search Engines
Introduction to architecture of search engines
Indexing
Introduction of information retrieval models
Web pages ranking algorithms
Methods for evaluating search engines
Launch a search engine using Opensource Tools

Research Area

• Information retrieval

Providers

• Rahman Jalayer

Present Type

Virtual

Holding Time

Date: October 23rd
Time: 2PM

Duration

4 hours

Files

link

Social Network Analysis Using Deep Learning Approach

Abstract

Nowadays, many people use social network’s features as one of the most popular tools of communication to communicate with each other, share news or their subjects of interests. A huge volume of interaction data created by users and their behavior has acted as motivation for researchers to discover knowledge. This information can be categorized into three categories: textual information, link information, and time information. Since most of the data created by users are text, textual data plays an important role in knowledge discovery from social networks. Text data in social networks are short and contains idioms, abbreviation and misspelled words. Also, users are not required to obey grammatical rules. Therefore, despite many advancements in NLP techniques, considering the points mentioned, processing text messages of a social network is known as a challenge because these techniques need manual feature extraction which is time consuming and usually results in an incomplete set of features. Deep learning methods have made feature extraction easy enough that these methods are commonly used in NLP. The purpose of this workshop is to introduce the social network and how to use deep learning in social networks’ textual data processing. This workshop is divided into two main parts. Headlines are listed below.

Headlines

Introduction to social network analysis
Role of social networks in user applications
Challenges of social networks' data processing and their solutions
social networks text mining
Use of deep learning to process textual data of social networks
Introduction to methods of deep learning
The implementation of an application program with the purpose of using deep learning tools to process the text data of Twitter social network

Research Area

• Social networks analysis

Providers

• Maryam Khodabakhsh
• Fattaneh Zarrin Kalam

Present Type

Virtual

Holding Time

Date: October 24th
Time: 8AM

Duration

4 hours

Files

link

Getting to know Microsoft’s business intelligence and data mining tools in SQL Server 2016

Abstract

The rapid change in the operations environment of companies and organizations drives them to use computer support for their operations. This means that pressure from outside of the firm leads to the creation of a response approach inside the organization which should be fast considering environment speed. This is due to the environment being very competitive and thus to have a quick response time, computer systems must come to firm’s assist. Computer systems can make the response process faster and easier. Business intelligence systems help managers and businesses in a variety of ways:
• They facilitate calculations and speed them up.
• They improve communication and cooperation.
• They increase the productivity of the people.
• They will enable and improve data management.
• They increase accessibility
These methods have resulted in fundamental decisions in businesses which will end up making decision support systems for organizations, firms, and businesses. The outlook of business intelligence can be described as follows: “Improving operations and key processes of a business through supplying information and crucial knowledge about business in the right time, right format and for all organizational levels”. The main purpose of business intelligence is to optimize, analyze, control and monitor business processes and operations. Following business intelligence purposes, three main functions in different businesses are created which are as follows:
• To create vision: make use of businesses’ past and present to predict the future.
• Management and analysis of business: define key performance indicators.
• Reporting: makes manual reports of every moment on different aspects of the system’s data.
Nowadays, data analysis is easy due to using tools and techniques of this field. One of the most effective methods of data analysis is data mining, which searches to discover knowledge from big databases in the form of patterns and probabilistic rules.

Headlines

Introduction to data warehouse and how it is created
A survey on ETL and how to use it (SSIS service)
Process of creating OLAP (SSAS service)
Process of visualization and data illustration (SSRS and Power BI services)
Running data mining algorithms in SQL Server 2016

Research Area

• Process mining

Providers

• Seyed Ahmad Tousi
• Ehsan Asgarian

Present Type

Virtual

Holding Time

Date: October 24th
Time: 2PM

Duration

4 hours

Files

link

Recommending Systems

Abstract

Nowadays, the importance of recommending systems are not overlooked. These systems are noticed due to their effective role in reducing information overhead and increasing sales and business boom, especially in E-commerce. The purpose of these systems is to predict the rating or priority that user will give to different items. These items can be movie, music, book, news, papers, commercial products and etc. these systems use users’ behavior analysis and items to recommend best items to users and helps their users to reach their target faster in a huge volume of data. In this workshop, we introduce different kinds of recommending systems and their uses in our environment, their common methods, their challenges, and their evaluation metrics

Headlines

Introduction to recommending systems
examples of commercial recommending systems in real world
types of recommending systems
introducing common methods and techniques in recommending systems
introducing the challenges in this area and the way to solve it
Methods for evaluating recommending systems
Tools for recommending systems

Research Area

• Information retrieval

Providers

• Fatemeh Pourgholamali
• Mahsa Khorasani

Present Type

Virtual

Holding Time

Date: October 25th
Time: 8AM

Duration

4 hours

Files

link