Fatih Cavdur

Neural network applications for automatic new topic identification

Purpose – This study aims to propose an artificial neural network to identify automatically topic changes in a user session by using the statistical characteristics of queries, such as time intervals and query reformulation patterns.Design/methodology/approach – A sample data log from the Norwegian search engine FAST (currently owned by Overture) is selected to train the neural network and then the neural network is used to identify topic changes in the data log.Findings – A total of 98.4 percent of topic shifts and 86.6 percent of topic continuations were estimated correctly.Originality/value – Content analysis of search engine user queries is an important task, since successful exploitation of the content of queries can result in the design of efficient information retrieval algorithms for search engines, which can offer custom‐tailored services to the web user. Identification of topic changes within a user search session is a key issue in the content analysis of search engine user queries.

Neural network applications for automatic new topic identification on excite web search engine data logs

The analysis of contextual information in search engine query logs is an important, yet difficult task. Users submit few queries, and search multiple topics sometimes with closely related context. Identification of topic changes within a search session is an important branch of contextual information analysis. The purpose of this study is to propose a topic identification algorithm using neural networks. A sample from the Excite data log is selected to train the neural network and then the neural network is used to identify topic changes in the data log. As a result, 76% of topic shifts and 92% of topic continuations are identified correctly.

Automatic new topic identification in search engine transaction logs

Purpose – Content analysis of search engine user queries is an important task, since successful exploitation of the content of queries can result in the design of efficient information retrieval algorithms of search engines, which can offer custom‐tailored services to the web user. Identification of topic changes within a user search session is a key issue in content analysis of search engine user queries. The purpose of this study is to address these issues.Design/methodology/approach – This study applies genetic algorithms and Dempster‐Shafer theory, proposed by He et al., to automatically identify topic changes in a user session by using statistical characteristics of queries, such as time intervals and query reformulation patterns. A sample data log from the Norwegian search engine FAST (currently owned by overture) is selected to apply Dempster‐Shafer theory and genetic algorithms for identifying topic changes in the data log.Findings – As a result, 97.7 percent of topic shifts and 87.2 percent of to...

A network view of business systems

In this paper, we present a novel business network construction approach, where the nodes of the network correspond to the names of the companies in a particular stock market index, and its links show the co-occurrence of two company names in daily news. Our approach consists of two phases, in which search for the company names in the news articles and network construction operations are performed, respectively. To increase the quality of results, each article is classified as business news or not business news before these operations, and only the articles that are classified as business news are considered for network construction. The resulting network presents a visualization of the business events and company relationships during the corresponding time period. We study both co-occurrences as well as single occurrences of company names in the articles scanned in our analysis.

Introducing the MCHF/OVRP/SDMP: Multicapacitated/Heterogeneous Fleet/Open Vehicle Routing Problems with Split Deliveries and Multiproducts

In this paper, we analyze a real-world OVRP problem for a production company. Considering real-world constrains, we classify our problem as multicapacitated/heterogeneous fleet/open vehicle routing problem with split deliveries and multiproduct (MCHF/OVRP/SDMP) which is a novel classification of an OVRP. We have developed a mixed integer programming (MIP) model for the problem and generated test problems in different size (10–90 customers) considering real-world parameters. Although MIP is able to find optimal solutions of small size (10 customers) problems, when the number of customers increases, the problem gets harder to solve, and thus MIP could not find optimal solutions for problems that contain more than 10 customers. Moreover, MIP fails to find any feasible solution of large-scale problems (50–90 customers) within time limits (7200 seconds). Therefore, we have developed a genetic algorithm (GA) based solution approach for large-scale problems. The experimental results show that the GA based approach reaches successful solutions with 9.66% gap in 392.8 s on average instead of 7200 s for the problems that contain 10–50 customers. For large-scale problems (50–90 customers), GA reaches feasible solutions of problems within time limits. In conclusion, for the real-world applications, GA is preferable rather than MIP to reach feasible solutions in short time periods.

Investigating the Performance of Automatic New Topic Identification Across Multiple Datasets

Recent studies on automatic new topic identification in Web search engine user sessions demonstrated that neural networks are successful in automatic new topic identification. However most of this work applied their new topic identification algorithms on data logs from a single search engine. In this study, we investigate whether the application of neural networks for automatic new topic identification are more successful on some search engines than others. Sample data logs from the Norwegian search engine FAST (currently owned by Overture) and Excite are used in this study. Findings of this study suggest that query logs with more topic shifts tend to provide more successful results on shift-based performance measures, whereas logs with more topic continuations tend to provide better results on continuation-based performance measures.

A two-phase binary-goal programming-based approach for optimal project-team formation

Abstract This study presents a two-phase binary-goal programming-based approach for solving a novel system design project-team formation problem which involves several restrictions and requirements as well as the preferences of the potential team members. The problem instance considered in this study basically includes two types of allocations, as the allocations of students and academic advisers, which are performed in the first and second phase of the proposed solution approach, respectively. Although it represents a particular case as represented in this study, it can be easily generalised to be used for solving similar project-team formation problems. We implement our methodology on a real-life problem in an academic institution and compare our solutions to the real-life allocations performed manually. It is noted that, in terms of satisfying the goals of the problem, our approach significantly outperforms the real-life allocations. In addition, computational results show our model’s ability to solve similar-sized real-life problems in reasonable time periods on an average personal computer, implying its potential for significant savings in terms of the human resources available.