Tweet Topic Classification Using Distributed Language Representations

Li, Quanzhi; Shah, Sameena; Liu, Xiaomo; Nourbakhsh, Armineh; Fang, Rui

doi:10.1109/wi.2016.0022

Cited by 16 publications

(13 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ying Quanzhi Li et al [3] present numerous tweet subject matter class techniques by way of exploiting different varieties of data: tweet text, tweet text plus entity know-how base, word embeddings derived from tweet text, distributed representations of tweets, and topical word. The word embedding, topical word embedding and sentence representation models are generated from billions of phrases from tweets without supervision.…”

Section: Literature Reviewmentioning

confidence: 99%

Untitled

2020

IJET

View full text Add to dashboard Cite

Twitter data accumulated so far make it in all likelihood to find out the distribution and go with the flow of mass tastes and opinions, which greatly help in product recommendation, target advertising and marketing and so on. This challenge proposes a subject model known as twitter hierarchical Latent Dirichlet Allocation (thLDA). Based on hierarchical latent Dirichlet allocation, thLDA aims to automatically mine the hierarchical measurement of tweets' subjects, which may be further employed for text Here, the phrases present in every of the topics with extra importance are extracted out and their beta value is found out. Furthermore, thLDA analyzes the relationships of words in tweets to get a more powerful measurement. Extensive experiments are carried out on Twitter information and the effectiveness of thLDA is evaluated. The consequences show that it outperforms properly amongst other current topic fashions in mining. This venture improves the topics mining amongst two subjects as well as three subjects. In addition, conditional chance is finished for Naïve Bayes Classification of important terms in the given statistics set so that the whole words possibilities in all the categories are found out and displayed. The task is designed the usage of R Studio 1.0. The coding language used is R 3.4.4.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Untitled

2020

IJET

View full text Add to dashboard Cite

show abstract

“…We performed the experimental evaluation using two different baselines, the first one based on tweet textual features only and the second one representing another alternative of contextual enrichment, namely word embeddings (KENTER;RIJKE, 2015;LI et al, 2016).…”

Section: List Of Figuresmentioning

confidence: 99%

A framework for event classification in tweets based on hybrid semantic enrichment

Romero

Becker

2019

Expert Systems with Applications

View full text Add to dashboard Cite

Social Media platforms have become key as a means of spreading information, opinions or awareness about real-world events. Twitter stands out due to the huge volume of messages about all sorts of topics posted every day. Such messages are an important source of useful information about events, presenting many useful applications (e.g. the detection of breaking news, real-time awareness, updates about events). However, text classification on Twitter is by no means a trivial task that can be handled by conventional Natural Language Processing techniques. In addition, there is no consensus about the definition of which kind of tasks are executed in the Event Identification and Classification in tweets, since existing approaches often focus on specific types of events, based on specific assumptions, which makes it difficult to reproduce and compare these approaches in events of distinct natures. In this work, we aim at building a unifying framework that is suitable for the classification of events of distinct natures. The framework has as key elements: a) external enrichment using related web pages for extending the conceptual features contained within the tweets; b) semantic enrichment using the Linked Open Data cloud to add related semantic features; and c) a pruning technique that selects the semantic features with discriminative potential. We evaluated our proposed framework using a broad experimental setting, that includes: a) seven target events of different natures; b) different combinations of the conceptual features proposed (i.e. entities, vocabulary and their combination); c) distinct feature extraction strategies (i.e. from tweet text and web related documents); d) different methods for selecting the discriminative semantic features (i.e. pruning, feature selection, and their combination); and e) two classification algorithms. We also compared the proposed framework against another kind of contextual enrichment based on word embeddings. The results showed the advantages of using the proposed framework, and that our solution is a feasible and generalizable method to support the classification of distinct event types.

show abstract

“…As far as a element group comes to a shape and don't bring a blend, it is said to be as an facts. This plan quicks up discovering and makes simpler for fixed interval group upgrade [14]. here computation method is used for data grouping and finding nearest neighbour search based event detection algorithm [15]…”

Section: B Event Clusteringmentioning

confidence: 99%

“…public network provides massive power to civilan news writers and evidence to extend data in case of events. The overview of investigation found that 10-20% of news firstly seen on Twitter [14]. Rather than going through all the news again and again, he can simply use this web application to notify him on his computer for the top trending and prevalent news of a specific period.…”

Section: Architecturementioning

confidence: 99%

Accept and Scale Traditional News Elements By Utilizing Public Network Elements

Rani*¹,

Srinivasu²

2019

IJITEE

View full text Add to dashboard Cite

Media orgins, specifically the news elements, have traditionally informed us incidents of daily life. In recent times, public network services such as Twitter provide an huge amount of numerous smaller collections generated from different user data. In this work, a new technique is proposed to filter noise and to obtain the data which are similar in news content which is considered to be valuable. It has a bottom-up approach to news recognition, and without of the help of an predefined origins of firm or topics. Rather, determines growing talks and messages daily to choose that are news-like. We introduce a hierarchical Bayesian model that jointly models the news media services and social media services and we show that our proposed model can capture different topics for individual datasets. In The proposed system the process is to change these instances into a regular model of characteristics and classes, System denote a oversee method based on a graphical paradigm to identify webbing poll show that the supervised method substantially improvize existing method. lastly the persons interactions while all the interests of the person in social media are considered hence the proposed work identifies and ranks news topic. In this paper we reveal our main domain, model and estimations are used to attain the goal, and instructions learnt parallely.

show abstract

Tweet Topic Classification Using Distributed Language Representations

Cited by 16 publications

References 14 publications

Untitled

Untitled

A framework for event classification in tweets based on hybrid semantic enrichment

Accept and Scale Traditional News Elements By Utilizing Public Network Elements

Contact Info

Product

Resources

About