User Query Enhancement for Behavioral Targeting

2017 ◽  
pp. 413-433
Author(s):  
Wei Xiong ◽  
Y. F. Brook Wu
Keyword(s):  
Topic Model ◽  
Classification Performance ◽  
Online Publishing ◽  
User Intent ◽  
User Classification ◽  
Behavioral Targeting ◽  
User Query ◽  
Publishing World ◽  
The Right ◽  
User Queries

Ad targeting has been receiving more and more attention in the online publishing world, where advertisers want their ads to be seen by potential consumers at the right time. This chapter aims to address the major challenges with user queries in the context of behavioral targeting advertising by proposing a user intent representation strategy and a query enhancement mechanism. The authors focus on investigating the intent based user classification performance and the effectiveness of user segmentation under a topic model that helps explore semantic relation between user queries in behavioral targeting. In addition, the authors propose an alternative to define user's search intent for the evaluation purpose, in the case that the dataset is sanitized.

Intent-Based User Segmentation with Query Enhancement

2013 ◽  
Vol 3 (4) ◽  
pp. 1-17
Author(s):  
Wei Xiong ◽  
Michael Recce ◽  
Brook Wu
Keyword(s):  
Clustering Algorithm ◽  
Topic Model ◽  
Online Advertising ◽  
Empirical Evaluation ◽  
Enhancement Mechanism ◽  
User Classification ◽  
User Behaviors ◽  
Segmentation Strategy ◽  
Behavioral Targeting ◽  
User Query

With the rapid advancement of the internet, accurate prediction of user's online intent underlying their search queries has received increasing attention from online advertising community. This paper aims to address the major challenges with user queries in the context of behavioral targeting advertising by proposing a query enhancement mechanism that augments user's queries by leveraging a user query log. The empirical evaluation demonstrates that the authors' methodology for query enhancement achieves greater improvement than the baseline models in both intent-based user classification and user segmentation. Different from traditional user segmentation methods, which take little semantics of user behaviors into consideration, the authors propose a novel user segmentation strategy by incorporating the query enhancement mechanism with a topic model to mine the relationships between users and their behaviors in order to segment users in a semantic manner. Comparing with a classical clustering algorithm, K-means, the experimental results indicate that the proposed user segmentation strategy helps improve behavioral targeting effectiveness significantly. This paper also proposes an alternative to define user's search intent for the evaluation purpose, in the case that the dataset is sanitized. This approach automatically labels users in a click graph, which are then used in training an intent-based user classifier.

EdgeDIPN

2020 ◽  
Vol 14 (3) ◽  
pp. 320-328
Author(s):  
Long Guo ◽  
Lifeng Hua ◽  
Rongfei Jia ◽  
Fei Fang ◽  
Binqiang Zhao ◽  
...  
Keyword(s):  
Real Time ◽  
Large Scale ◽  
Attention Mechanism ◽  
Data Sources ◽  
User Intent ◽  
Multiple User ◽  
Shopping Experience ◽  
Data Source ◽  
Intent Prediction ◽  
The Right

With the rapid growth of e-commerce in recent years, e-commerce platforms are becoming a primary place for people to find, compare and ultimately purchase products. To improve online shopping experience for consumers and increase sales for sellers, it is important to understand user intent accurately and be notified of its change timely. In this way, the right information could be offered to the right person at the right time. To achieve this goal, we propose a unified deep intent prediction network, named EdgeDIPN, which is deployed at the edge, i.e., mobile device, and able to monitor multiple user intent with different granularity simultaneously in real-time. We propose to train EdgeDIPN with multi-task learning, by which EdgeDIPN can share representations between different tasks for better performance and saving edge resources in the meantime. In particular, we propose a novel task-specific attention mechanism which enables different tasks to pick out the most relevant features from different data sources. To extract the shared representations more effectively, we utilize two kinds of attention mechanisms, where the multi-level attention mechanism tries to identify the important actions within each data source and the inter-view attention mechanism learns the interactions between different data sources. In the experiments conducted on a large-scale industrial dataset, EdgeDIPN significantly outperforms the baseline solutions. Moreover, EdgeDIPN has been deployed in the operational system of Alibaba. Online A/B testing results in several business scenarios reveal the potential of monitoring user intent in real-time. To the best of our knowledge, EdgeDIPN is the first full-fledged real-time user intent understanding center deployed at the edge and serving hundreds of millions of users in a large-scale e-commerce platform.

Online-ready? Factor structure and internal consistency of a scale to measure students’ online journalism self-efficacy

2021 ◽  
Vol 00 (00) ◽  
pp. 1-25
Author(s):  
Joseph Njuguna
Keyword(s):  
Social Media ◽  
Factor Structure ◽  
Self Efficacy ◽  
Mass Communication ◽  
Online News ◽  
Media Industry ◽  
Online Journalism ◽  
Online Publishing ◽  
Online Skills ◽  
The Right

With the integration of social media in journalism practice, media training institutions must focus on preparing future media professionals with the right mix of digital skills for the industry. Although efforts to improve students’ online skills readiness are evident in schools, no reliable tool exists to assess students’ confidence in doing online journalism tasks upon graduation. This study develops and validates an Online Journalism Self-Efficacy Scale (OJSES) that can be used to measure mass communication students’ perceptions of their self-efficacy for online journalism work. Items for the proposed scale were developed from a comprehensive literature review and refined by eight online journalism professionals (five online journalism lecturers and three online news editors). To explore the factor structure of the tool, exploratory factor analysis of data from a sample of finalist undergraduate mass communication students (n = 182) in five Rwandan universities was done. Results suggested that the OJSES is a five-dimensional tool that comprises 27 items. This scale measures online journalism self-efficacy in terms of students’ capabilities to conduct online journalism research, communicate with social media tools, create and share multimedia content online, observe ethical online publishing and use social media to solve organizational problems. The scale demonstrated reliability with a Cronbach’s alpha value of 0.785 and the five self-efficacy dimensions explaining 51.1 per cent of the total variance. The scale’s psychometric soundness implied its suitability not only to empirically measure the students’ confidence in working in online environments but also guide capacity-building for the required online skills for the media industry.

Topic-BERT: Detecting harmful information from social media

2021 ◽  
pp. 1-10
Author(s):  
Wang Gao ◽  
Hongtao Deng ◽  
Xun Zhu ◽  
Yuan Fang
Keyword(s):  
Social Media ◽  
Language Processing ◽  
Topic Model ◽  
Classification Performance ◽  
Critical Research ◽  
Short Text ◽  
Additional Information ◽  
Proposed Model ◽  
Weight Calculation ◽  
Two Stages

Harmful information identification is a critical research topic in natural language processing. Existing approaches have been focused either on rule-based methods or harmful text identification of normal documents. In this paper, we propose a BERT-based model to identify harmful information from social media, called Topic-BERT. Firstly, Topic-BERT utilizes BERT to take additional information as input to alleviate the sparseness of short texts. The GPU-DMM topic model is used to capture hidden topics of short texts for attention weight calculation. Secondly, the proposed model divides harmful short text identification into two stages, and different granularity labels are identified by two similar sub-models. Finally, we conduct extensive experiments on a real-world social media dataset to evaluate our model. Experimental results demonstrate that our model can significantly improve the classification performance compared with baseline methods.

Artificial Odour Classification System

2018 ◽  
pp. 25-37
Author(s):  
Nor Idayu Mahat ◽  
Maz Jamilah Masnan ◽  
Ali Yeon Md Shakaff ◽  
Ammar Zakaria ◽  
Muhd Khairulzaman Abdul Kadir
Keyword(s):  
Feature Selection ◽  
Mahalanobis Distance ◽  
Feature Selection Method ◽  
Classification Performance ◽  
Good Choice ◽  
Classification Rule ◽  
Good Classification Performance ◽  
The Right ◽  
The Cost

This chapter overviews the issue of multicollinearity in electronic nose (e-nose) classification and investigates some analytical solutions to deal with the problem. Multicollinearity effect may harm classification analysis from producing good parameters estimate during the construction of the classification rule. The common approach to deal with multicollinearity is feature extraction. However, the criterion used in extracting the raw features based on variances may not be appropriate for the ultimate goal of classification accuracy. Alternatively, feature selection method would be advisable as it chooses only valuable features. Two distance-based criteria in determining the right features for classification purposes, Wilk's Lambda and bounded Mahalanobis distance, are applied. Classification with features determined by bounded Mahalanobis distance statistically performs better than Wilk's Lambda. This chapter suggests that classification of e-nose with feature selection is a good choice to limit the cost of experiments and maintain good classification performance.

Application of Genetic Algorithm and Back Propagation Neural Network for Effective Personalize Web Search-Based on Clustered Query Sessions

2016 ◽  
Vol 7 (1) ◽  
pp. 33-49 ◽  
Author(s):  
Suruchi Chawla
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Web Search ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Information Need ◽  
Data Set ◽  
User Query ◽  
User Queries

In this paper novel method is proposed using hybrid of Genetic Algorithm (GA) and Back Propagation (BP) Artificial Neural Network (ANN) for learning of classification of user queries to cluster for effective Personalized Web Search. The GA- BP ANN has been trained offline for classification of input queries and user query session profiles to a specific cluster based on clustered web query sessions. Thus during online web search, trained GA –BP ANN is used for classification of new user queries to a cluster and the selected cluster is used for web page recommendations. This process of classification and recommendations continues till search is effectively personalized to the information need of the user. Experiment was conducted on the data set of web user query sessions to evaluate the effectiveness of Personalized Web Search using GA optimized BP ANN and the results confirm the improvement in the precision of search results.

scholarly journals Parkinson's Detection Based On Combined CNN And LSTM Using Enhanced Speech Signals With Variational Mode Decomposition

2021 ◽  
Author(s):  
Mehmet Bilal ER ◽  
Esme ISIK ◽  
Ibrahim ISIK
Keyword(s):  
Network Architecture ◽  
Short Term Memory ◽  
Early Period ◽  
Classification Performance ◽  
Speech Disorders ◽  
Speech Signals ◽  
Variational Mode Decomposition ◽  
Mode Decomposition ◽  
Deep Networks ◽  
The Right

Abstract The dysfunction of the cells in the brain that contain the substance known as dopamine, which enables brain cells to interact with each other, results in Parkinson's disease (PD). PD can cause many non-motor and motor symptoms such as speech and smell. One of the difficulties that Parkinson’s patients can experience is a change in speech or speaking difficulties. Therefore, the right diagnosis in the early period is important in reducing the possible effects of speech disorders caused by the disease. Speech signal of Parkinson patients shows major differences compared to normal people. In this study, a new approach based on pre-trained deep networks and Long short-term memory (LSTM) by using mel-spectrograms obtained from denoised speech signals with Variational Mode Decomposition (VMD) for detecting PD from speech sounds is proposed. The proposed model consists of four stages. In the first step, the noise is removed by applying VMD to the signals. In the second stage, Mel-spectrograms are extracted from the enhanced sound signals with VMD. In the third stage, pre-trained deep networks are preferred to extract deep features from the Mel-spectrograms. For this purpose, ResNet-18, ResNet-50 and ResNet-101 models are used as pre-trained deep network architecture. In the last step, the classification process is occured by giving these features as input to the LSTM model, which is designed to define sequential information from the extracted features. Experiments are performed with the PC-GITA dataset, which consists of two classes and is widely used in the literature. The results obtained from the proposed method are compared with the latest methods in the literature, it is seen that it has a better performance in terms of classification performance.

scholarly journals A New Vector Representation of Short Texts for Classification

2019 ◽  
Vol 17 (2) ◽  
pp. 241-249
Author(s):  
Yangyang Li ◽  
Bo Liu
Keyword(s):  
Text Classification ◽  
Web Search ◽  
Latent Dirichlet Allocation ◽  
Topic Model ◽  
Classification Performance ◽  
New Method ◽  
Data Sets ◽  
Text Data ◽  
Short Text ◽  
Space Model

Short and sparse characteristics and synonyms and homonyms are main obstacles for short-text classification. In recent years, research on short-text classification has focused on expanding short texts but has barely guaranteed the validity of expanded words. This study proposes a new method to weaken these effects without external knowledge. The proposed method analyses short texts by using the topic model based on Latent Dirichlet Allocation (LDA), represents each short text by using a vector space model and presents a new method to adjust the vector of short texts. In the experiments, two open short-text data sets composed of google news and web search snippets are utilised to evaluate the classification performance and prove the effectiveness of our method.

Politics and Book Publishing in Argentina

2020 ◽  
Author(s):  
Ezequiel Saferstein
Keyword(s):  
Publishing House ◽  
Book Publishing ◽  
Decisive Role ◽  
Cultural Economic ◽  
Cultural Objects ◽  
Historical Significance ◽  
Political Ideas ◽  
Publishing World ◽  
Production And Distribution ◽  
The Right

In a country where literacy rates are among the highest in the region, books are cultural objects cherished by vast sectors of the Argentine population as well as powerful symbolic, cultural, economic, and political artefacts. In particular, books on politics are an indispensable segment in the catalog of any Argentine publishing house. The vertiginous nature of politics and the historical significance of the book in Argentine society are such that the publishing sector has been—and still remains—one of the preferred spaces where symbolic and political power is disputed. Throughout the 20th century and the first two decades of the 21st century, the publishing market responded to different historical circumstances by producing headlines that sought to engage readers in different ways, helping them make life choices and understand the significance of their own time, as well as forming or reinforcing their opinions. Manufactured from the Left to the Right, books on politics expressed and shaped wills and aspirations, serving as combat weapons and means for the creation of spaces where ideas and political sentiments flourish. There are historical ties between the Argentinean publishing and political spheres, and the publishing process works as a fundamental form of mediation concerning the production and distribution of political ideas. Against the image of the book as an exclusive bridge connecting the authors with the reading public, a sociological and material viewpoint might focus on the publishing world and its protagonists: the ghost editors and agents who play an indispensable and decisive role in the processes whereby a book becomes an entitled cultural, economic, and political intervention—a great factory of ideas, discourses, and products with material and symbolic ramifications that influence public debates and agendas.

Sign in / Sign up

Export Citation Format

Share Document