scholarly journals Fast and Effective Retrieval for Large Multimedia Collections

2021 ◽  
Vol 5 (3) ◽  
pp. 33
Author(s):  
Stefan Wagenpfeil ◽  
Binh Vu ◽  
Paul Mc Kevitt ◽  
Matthias Hemmje
Keyword(s):  
Social Media ◽  
Feature Detection ◽  
Feature Fusion ◽  
Large Data ◽  
Level Of Detail ◽  
Multimedia Big Data ◽  
Levels Of Detail ◽  
Indexing And Retrieval ◽  
Similarity Algorithm ◽  
High Level

The indexing and retrieval of multimedia content is generally implemented by employing feature graphs. These graphs typically contain a significant number of nodes and edges to reflect the level of detail in feature detection. A higher level of detail increases the effectiveness of the results, but also leads to more complex graph structures. However, graph traversal-based algorithms for similarity are quite inefficient and computationally expensive, especially for large data structures. To deliver fast and effective retrieval especially for large multimedia collections and multimedia big data, an efficient similarity algorithm for large graphs in particular is desirable. Hence, in this paper, we define a graph projection into a 2D space (Graph Code) and the corresponding algorithms for indexing and retrieval. We show that calculations in this space can be performed more efficiently than graph traversals due to the simpler processing model and the high level of parallelization. As a consequence, we demonstrate experimentally that the effectiveness of retrieval also increases substantially, as the Graph Code facilitates more levels of detail in feature fusion. These levels of detail also support an increased trust prediction, particularly for fused social media content. In our mathematical model, we define a metric triple for the Graph Code, which also enhances the ranked result representations. Thus, Graph Codes provide a significant increase in efficiency and effectiveness, especially for multimedia indexing and retrieval, and can be applied to images, videos, text and social media information.

Vivid Imagers Are Better at Detecting Salient Changes

2006 ◽  
Vol 27 (4) ◽  
pp. 218-228 ◽  
Author(s):  
Paul Rodway ◽  
Karen Gillies ◽  
Astrid Schepman
Keyword(s):  
Individual Differences ◽  
Change Detection ◽  
Visual Imagery ◽  
Detection Task ◽  
Level Of Detail ◽  
Detection Accuracy ◽  
Change Detection Task ◽  
Term Change ◽  
High Level

This study examined whether individual differences in the vividness of visual imagery influenced performance on a novel long-term change detection task. Participants were presented with a sequence of pictures, with each picture and its title displayed for 17  s, and then presented with changed or unchanged versions of those pictures and asked to detect whether the picture had been changed. Cuing the retrieval of the picture's image, by presenting the picture's title before the arrival of the changed picture, facilitated change detection accuracy. This suggests that the retrieval of the picture's representation immunizes it against overwriting by the arrival of the changed picture. The high and low vividness participants did not differ in overall levels of change detection accuracy. However, in replication of Gur and Hilgard (1975) , high vividness participants were significantly more accurate at detecting salient changes to pictures compared to low vividness participants. The results suggest that vivid images are not characterised by a high level of detail and that vivid imagery enhances memory for the salient aspects of a scene but not all of the details of a scene. Possible causes of this difference, and how they may lead to an understanding of individual differences in change detection, are considered.

scholarly journals MTI-YOLO: A Light-Weight and Real-Time Deep Neural Network for Insulator Detection in Complex Aerial Images

Energies ◽  
2021 ◽  
Vol 14 (5) ◽  
pp. 1426
Author(s):  
Chuanyang Liu ◽  
Yiquan Wu ◽  
Jingjing Liu ◽  
Jiaming Han
Keyword(s):  
Feature Detection ◽  
Feature Fusion ◽  
Memory Storage ◽  
Aerial Images ◽  
Detection Accuracy ◽  
Composite Insulator ◽  
Running Time ◽  
Scale Feature ◽  
Multi Scale ◽  
Good Trade

Insulator detection is an essential task for the safety and reliable operation of intelligent grids. Owing to insulator images including various background interferences, most traditional image-processing methods cannot achieve good performance. Some You Only Look Once (YOLO) networks are employed to meet the requirements of actual applications for insulator detection. To achieve a good trade-off among accuracy, running time, and memory storage, this work proposes the modified YOLO-tiny for insulator (MTI-YOLO) network for insulator detection in complex aerial images. First of all, composite insulator images are collected in common scenes and the “CCIN_detection” (Chinese Composite INsulator) dataset is constructed. Secondly, to improve the detection accuracy of different sizes of insulator, multi-scale feature detection headers, a structure of multi-scale feature fusion, and the spatial pyramid pooling (SPP) model are adopted to the MTI-YOLO network. Finally, the proposed MTI-YOLO network and the compared networks are trained and tested on the “CCIN_detection” dataset. The average precision (AP) of our proposed network is 17% and 9% higher than YOLO-tiny and YOLO-v2. Compared with YOLO-tiny and YOLO-v2, the running time of the proposed network is slightly higher. Furthermore, the memory usage of the proposed network is 25.6% and 38.9% lower than YOLO-v2 and YOLO-v3, respectively. Experimental results and analysis validate that the proposed network achieves good performance in both complex backgrounds and bright illumination conditions.

scholarly journals Predicting Implicit User Preferences with Multimodal Feature Fusion for Similar User Recommendation in Social Media

2021 ◽  
Vol 11 (3) ◽  
pp. 1064
Author(s):  
Jenq-Haur Wang ◽  
Yen-Tsang Wu ◽  
Long Wang
Keyword(s):  
Social Media ◽  
Feature Fusion ◽  
Relevant Information ◽  
Image Features ◽  
User Preferences ◽  
User Preference ◽  
Late Fusion ◽  
Multimodal Features ◽  
Fusion Methods ◽  
Text Features

In social networks, users can easily share information and express their opinions. Given the huge amount of data posted by many users, it is difficult to search for relevant information. In addition to individual posts, it would be useful if we can recommend groups of people with similar interests. Past studies on user preference learning focused on single-modal features such as review contents or demographic information of users. However, such information is usually not easy to obtain in most social media without explicit user feedback. In this paper, we propose a multimodal feature fusion approach to implicit user preference prediction which combines text and image features from user posts for recommending similar users in social media. First, we use the convolutional neural network (CNN) and TextCNN models to extract image and text features, respectively. Then, these features are combined using early and late fusion methods as a representation of user preferences. Lastly, a list of users with the most similar preferences are recommended. The experimental results on real-world Instagram data show that the best performance can be achieved when we apply late fusion of individual classification results for images and texts, with the best average top-k accuracy of 0.491. This validates the effectiveness of utilizing deep learning methods for fusing multimodal features to represent social user preferences. Further investigation is needed to verify the performance in different types of social media.

Countering Threats

2014 ◽  
Vol 4 (2) ◽  
pp. 35-45
Author(s):  
Margarita Jaitner
Keyword(s):  
Social Media ◽  
Law Enforcement ◽  
Ad Hoc ◽  
Physical Violence ◽  
Deep Understanding ◽  
Current Status ◽  
Mobile Web ◽  
Use Of Social Media ◽  
Legal Frameworks ◽  
High Level

The increased adoption of social media has presented security and law enforcement authorities with significant new challenges. For example, the Swedish Security Service (SÄPO) asserts that a large proportion of radicalization takes place in open fora online. Still, approaches to contain social media-driven challenges to security, particularly in democratic societies, remain little explored. Nonetheless, this type of knowledge may become relevant in European countries in the near future: Amongst other factors, the challenging economic situation has resulted in increased public discontent leading to emergence or manifestation of groups that seek to challenge the existing policies by almost any means. Use of social media multiplies the number of vectors that need law enforcement attention. First, a high level of social media adaption allows groups to reach and attract a wider audience. Unlike previously, many groups today consist of a large but very loosely connected network. This lack of cohesion can present a challenge for authorities, to identify emerging key actors and assess threat levels. Second, a high level of mobile web penetration has allowed groups to ad-hoc organize, amend plans and redirect physical activities. Third, the tool social media is as not exclusive to potential perpetrators of unlawful action, but is as well available to law enforcement authorities. Yet, efficient utilization of social media requires a deep understanding of its nature and a well-crafted, comprehensive approach. Acknowledging the broad functionality of social media, as well as its current status in the society, this article describes a model process for security authorities and law enforcement work with social media in general and security services work in particular. The process is cyclic and largely modular. It provides a set of goals and tasks for each stage of a potential event, rather than fixed activities. This allows authorities to adapt the process to individual legal frameworks and organization setups. The approach behind the process is holistic where social media is regarded as both source and destination of information. Ultimately, the process aims at efficiently and effectively mitigating the risk of virtual and physical violence.

Conspicuous consumption of luxury experiences: an experimental investigation of status perceptions on social media

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Carolin Siepmann ◽  
Lisa Carola Holthoff ◽  
Pascal Kowalczuk
Keyword(s):  
Social Media ◽  
Personal Characteristics ◽  
Conspicuous Consumption ◽  
Individual Characteristics ◽  
Luxury Goods ◽  
Content Type ◽  
The Status ◽  
Self Actualization ◽  
The Individual ◽  
High Level

Purpose As luxury goods are losing their importance for demonstrating status, wealth or power to others, individuals are searching for alternative status symbols. Recently, individuals have increasingly used conspicuous consumption and displays of experiences on social media to obtain affirmation. This study aims to analyze the effects of luxury and nonluxury experiences, as well as traditional luxury goods on status- and nonstatus-related dimensions. Design/methodology/approach After presenting the theoretical foundation, the authors conduct a study with 599 participants to compare status perceptions elicited by the conspicuous consumption of luxury goods, luxury experiences and nonluxury experiences. The authors investigate whether experiences that are visibly consumed on Instagram are replacing traditional luxury goods as the most important status symbols. Furthermore, the authors examine the effects of the content shown on nonstatus-related dimensions and analyze whether status perceptions differ between female and male social media communicators. Finally, the authors analyze how personal characteristics (self-esteem, self-actualization and materialism) influence the status perceptions of others on social media. Findings The results show that luxury goods are still the most important means of displaying status. However, especially for women, luxury experiences are also associated with a high level of social status. Thus, the results imply important gender differences in the perceptions of status- and nonstatus-related dimensions. Furthermore, the findings indicate that, in particular, the individual characteristics of self-actualization and materialism affect status perceptions depending on the posted content. Originality/value While the research has already considered some alternative forms of conspicuous consumption, little attention has been given to experiences as status symbols. However, with their growing importance as substitutes for luxury goods and the rise of social media, the desire to conspicuously consume experiences is increasing. The authors address this gap in the literature by focusing on the conspicuous display of luxury and nonluxury experiences on social media.

scholarly journals Automated monitoring system for events detection in sewer network by distribution temperature sensing data measurement

2018 ◽  
Vol 78 (7) ◽  
pp. 1499-1508
Author(s):  
Abdelhak Kessili ◽  
Jes Vollertsen ◽  
Asbjørn Haaning Nielsen
Keyword(s):  
Large Data ◽  
Temperature Sensing ◽  
Wastewater Discharge ◽  
Sewer Pipe ◽  
Temperature Changes ◽  
Set Up ◽  
Comparison Of The Results ◽  
Events Detection ◽  
High Level ◽  
Automated Tool

Abstract This study is related to distribution temperature sensing (DTS) in sewers for tracing illicit or unintended inflows to foul sewers. A DTS measurement is performed with a fiber optic cable that is installed at the invert of a sewer pipe in combination with a standalone laser/computer instrument. This set-up generates in-sewer temperature measurements with high resolutions in time (every minute) and space (every metre) along the cable over long periods of time (weeks on end). The prolonged monitoring period in combination with the high level of detail in the dataset allows the study of anomalies (i.e., unexpected temperatures and/or temperature variations at certain locations), even if these only occur very infrequently. The objective of this paper is to develop an automated tool to analyze the large data masses and identify anomalies caused by illicit or unintended inflows. In this study, an algorithm for detecting the temperature changes that are caused by both wastewater discharge and inflow of stormwater are developed. A comparison of the results of the automated procedure to the results of a manual assessment of the datasets (Elmehaven, Denmark) shows that the automated procedure performs very well.

scholarly journals Intention to Online Shopping on Social Media

2021 ◽  
Vol 2 (2) ◽  
pp. 119
Author(s):  
Anita Akhirruddin
Keyword(s):  
Social Media ◽  
Online Shopping ◽  
Product Information ◽  
High Growth ◽  
Quality Of Information ◽  
Social Media Site ◽  
The Social ◽  
Quality Of Products ◽  
High Level

high growth of online shopping in Indonesia gave rise to many onlien websites and platforms in Indonesia. Facebook, which is one of the social networks that many people use around the world, is one of the online selling media that is in great demand because it can reach more people. Shopping online on Facebook in addition to providing benefits for sellers and buyers. Online shopping on Facebook requires a high level of trust from buyers regarding the quality of products and serv ices, and ease in obtaining product information and payments because there are no guarantees such as online shop platforms such as shoope, tokopedia, lazada and others which before the goods are received by the customer, then the money from the buyer can not be disbursed. So researchers are interested in researching online shopping interests on the social media site facebook. The results obtained are variable trust, ease of transaction and quality of information positively affect the interest in buying online on facebook.

scholarly journals Pre-university informal engagement with mathematical activities and the decision to study mathematics at university

2020 ◽  
Vol 18 (3) ◽  
pp. 10-22
Author(s):  
Katie Steckles ◽  
Peter Rowlett ◽  
Angharad Ugonna
Keyword(s):  
Social Media ◽  
Career Choice ◽  
Small Sample ◽  
First Year ◽  
Mathematical Activity ◽  
Online Videos ◽  
High Level ◽  
Watching Tv ◽  
Mathematical Activities

A survey was created to investigate the experiences of mathematics undergraduates with informal mathematical activity prior to starting university, and links these with the decision to study mathematics. A questionnaire was completed by a small sample of first-year undergraduates at two UK universities. Generally, incoming undergraduates are shown to have a high level of enjoyment of mathematics and engagement with informal mathematical activity. Popular activities included mathematical puzzles and games, and online videos about maths. Students were often engaged with family or via social media, playing computer, tablet or phone games, watching TV game shows with mathematical aspects and participating in organised competitions. Only around half engaged via talks or workshops organised through school and watching more structured documentaries or videos of lectures. Few participated in organised clubs. It seems there was greater engagement with ‘fun’ aspects of mathematics than with activities which demonstrate mathematics linked to career choice. The link to goals of outreach and similar initiatives is discussed, with further research indicated.

Sign in / Sign up

Export Citation Format

Share Document