1,720,964 research outputs found
Using gravitational force in terrain optimization problems
No full textOptimization problems represent algorithms designed for difficult problems which may require huge amount of space or computational time. Such kind of algorithms bring out solutions which are optimal and at the same time closer to the real life environments
where not everything is precise and the possibility of errors is present at every instant. Finding the minimal point in a terrain is a challenge of its own, especially when we are dealing with an unknown area. In order to tackle this problem, we thought
of making use of gravitational force, since it is proportionally related to earth center proximity. In an unknown terrain, we spread our agents that are capable to communicate information to one another at randomly generated positions. Later on, each of these agents calculates the gravity variation with altitude at its respective position. Since, we were looking to find the optimal minimum point in the terrain, after the gravity variation with altitude is computed by each agent, the highest gravity is found. This is communicated to the other agents as well and they start moving toward the agent that is currently found at an area with high gravity variation. The agents move toward the high gravity agent with a certain heuristic coefficient. During their path they may encounter other terrain points where the gravitational force is stronger, which would cause a change in the path of other agents making them move toward the newly found position. This is done until an optimal minimum is found by the agents. Our test results so far have been very promising. We aim to develop the algorithm furthermore in order to increase its efficiency and efficacy. We strongly believe that such an algorithm can be used to reach in the unexplored areas of ocean floor, or searching for minerals by minimizing the area of search in an optimal timeOptimization problems represent algorithms designed for difficult problems which may require huge amount of space or computational time. Such kind of algorithms bring out solutions which are optimal and at the same time closer to the real life environments
where not everything is precise and the possibility of errors is present at every instant. Finding the minimal point in a terrain is a challenge of its own, especially when we are dealing with an unknown area. In order to tackle this problem, we thought
of making use of gravitational force, since it is proportionally related to earth center proximity. In an unknown terrain, we spread our agents that are capable to communicate information to one another at randomly generated positions. Later on, each of these agents calculates the gravity variation with altitude at its respective position. Since, we were looking to find the optimal minimum point in the terrain, after the gravity variation with altitude is computed by each agent, th
Dynamic virtual bats algorithm (DVBA) for minimization of supply chain cost with embedded risk
No full textDynamic Virtual Bats Algorithm (DVBA) is a new optimization algorithm, which is tested on several benchmark functions for global optimization. However it has not been tested on a real world problem yet. In this paper DVBA has been applied to minimize the supply chain cost with other well known algorithms, Particle Swarm Optimization (PSO), Bat Algorithm (BA), Genetic Algorithm (GA) and Tabu Search (TS). Optimization of supply chain is considered as a real challenge by researchers because of its complexity. Big number of parameters to be controlled and their distributions, interconnections between parameters and dynamism are the main factors that increase the complexity of a supply chain. The result of the case study showed that the DVBA is much superior to other algorithms in terms of accuracy and efficiency. © 2014 IEEE
The Computational Psychology of Digital Shop Assistants
No full textThis proposal describes a project lying at the intersection of Computer Science and Social Sciences whose goal is to state and investigate some basic questions concerning recommender systems (henceforth RS’s). The advent of the internet has changed cultural markets in profound ways. The global volume of online purchases of music, books, movies, video games and other forms of cultural products has reached the 1.5 trillion dollars mark in 2014, and the trend is increasing. Today, an estimated 1.22 billions people acquire cultural products through the internet. RS’s are a key component of these online markets. In the old days, a regular customer of, say, a music shop, could get the advice of a knowledgeable shop assistant with whom s/he had developed a relationship of trust. Based on the knowledge of the customer’s taste and of the music world, the assistant could offer insightful suggestions to the customer, providing useful advice. Roughly speaking, a RS is a digital, algorithmic analogue of the shop assistant that, on the basis of the past online behaviour of the current customer and of the entire collective behaviour of online visitors, helps navigate the huge catalogue of online choices by providing suggestions in a purely algorithmic fashion. Thus, a visitor to the YouTube home site will be presented with a list of videos that, hopefully, will match his/her interests, and a person looking for a book on Amazon will likewise see a list of other interesting books to buy. In spite of the fact that RS’s are fundamental actors of online cultural markets, their power to shape and influence is still largely unknown. The goal of this proposal is to investigate the extent to which a cultural market can be affected by RS’s and the interplay between computational and psychological mechanisms underlying them
Songs of a Future Past - An Experimental Study of Online Persuaders
No full textIn this paper, we present the results of an extensive experimental study on users decisions inside an online setting. In the experiment, participants purchase songs using real money while having enough time to explore them at leisure before buying. In such a set up, surpisingly, common social influence signals such as star ratings, download counts and recommendations had no influence. However, as soon as the exploration was made slightly more cumbersome market inequality appeared. This is an indication that it is decision-making shortcuts, rather then social influence, to trigger distorting market effects
On the distortion of locality sensitive hashing
Get PDFGiven a notion of pairwise similarity between objects, locality sensitive hashing (LSH) aims to construct a hash function family over the universe of objects such that the probability two objects hash to the same value is their similarity. LSH is a powerful algorithmic tool for large scale applications and much work has been done to understand LSHable similarities, i.e., similarities that admit an LSH. In this paper we focus on similarities that are provably non-LSHable and propose a notion of distortion to capture the approximation of such a similarity by an LSHable similarity. We consider several well-known non-LSHable similarities and show tight upper and lower bounds on their distortion
The Distortion of Locality Sensitive Hashing
Get PDFGiven a pairwise similarity notion between objects, locality sensitive hashing (LSH) aims to construct a hash function family over the universe of objects such that the probability two objects hash to the same value is their similarity. LSH is a powerful algorithmic tool for large-scale applications and much work has been done to understand LSHable similarities, i.e., similarities that admit an LSH.
In this paper we focus on similarities that are provably non-LSHable and propose a notion of distortion to capture the approximation of such a similarity by a similarity that is LSHable. We consider several well-known non-LSHable similarities and show tight upper and lower bounds on their distortion. We also experimentally show that our upper bounds translate to
The limits of popularity-based recommendations, and the role of social ties
Get PDFIn this paper we introduce a mathematical model that captures some of the salient features of recommender systems that are based on popularity and that try to exploit social ties among the users. We show that, under very general conditions, the market always converges to a steady state, for which we are able to give an explicit form. Thanks to this we can tell rather precisely how much a market is altered by a recommendation system, and determine the power of users to influence others. Our theoretical results are complemented by experiments with real world social networks showing that social graphs prevent large market distortions in spite of the presence of highly influential user
Modeling a Career Office Information System with UML
Get PDFDifficulties faced by students while trying to penetrate into the labor market and challenges faced by university career offices when helping students build successful professional careers demand the building of mediums which would facilitate the transition of students to the market by bringing career office staff, students and businesses altogether in a single platform. This paper focuses on analysis and design based on the Unified Modeling Language (UML) of a Career Office Information System (COIS), whose aim is to establish connections between students and businesses, and simply the enormous work of a Career Office in a certain university
Together We Buy, Alone I Quit. Some Experimental Studies of Online Persuaders
No full textWe present a simple web experiment in which participants are asked to listen to a small number of songs and download the two they liked the most. In the experiment, participants were subject to common types of online feedbacks such as star ratings, recommendations and expert advice. Somewhat surprisingly, such online cues had no impact on market shares, but a significant difference emerged as far as market volume was concerned. When operating under the influence of online cues conjuring the presence of others activities soared: participants downloaded, listened to and rated songs much more than in the other scenarios
- …
