Find Jobs
Hire Freelancers

Data Processing, Analysis and Prediction.

$250-750 USD

已关闭
已发布将近 9 年前

$250-750 USD

货到付款
This project entails creating a classifier with the ability to predict when a news article published on the web is going to be breaking-news, that is, an article of great importance/interest, or not. For the purpose of this project you will be provided with a .csv data set that contains news article clusters for a certain time period. Each cluster contains an original news article by some news source (e.x CNN) identified by the fact that it has the oldest date value among other articles in the same cluster, and, republished versions of the original news article by other news sources (e.x BBC republished the CNN news article). The columns contained in the csv file are the following: id, url, date, categoryId, siteId, clusterId, count([login to view URL])
项目 ID: 7514241

关于此项目

12提案
远程项目
活跃9 年前

想赚点钱吗?

在Freelancer上竞价的好处

设定您的预算和时间范围
为您的工作获得报酬
简要概述您的提案
免费注册和竞标工作
12威客以平均价$919 USD来参与此工作竞价
用户头像
A proposal has not yet been provided
$555 USD 在10天之内
4.9 (140条评论)
6.9
6.9
用户头像
Hi, I can do this for you. Please send a massage in the PMB for details.......Best Regards flashsaiful
$555 USD 在10天之内
4.9 (111条评论)
6.5
6.5
用户头像
Hi, I have done several data mining projects where I did analysis of given data set. For finding published news with greatest importance that what I understand from given detail can be done by counting the republish time for one news id. But as you gave data set attributes how we can find the particular news if it republish; here I mean there should be unique News-Id that assign to each unique news and when it republish by another channel then by that news-Id with different URL( if url of news channel) or with different site-Id( if site-Id is of different channel) we can determine the republish count for each news and by that count we can determine article importance. Or can you please explain on which criteria any news becomes breaking news. what information is given in this data set for finding this. Looking forward to clarify above points. Regards, Rabia
$444 USD 在10天之内
5.0 (30条评论)
5.6
5.6
用户头像
Hi there, I am an expert on Big Data algorithm programming namely map-reduce, Hive Queries and Pig scripts. Hence I am confident that I will be able to predict the probability of a news article becoming a breaking-news or not. However based on my past experiences I have the following questions :- 1) Do you want any specified type of tool to be used like Apache Mahout ? 2) The classifier logic has to be developed from scratch by me or you already have it ready and I merely need to implement it in your dataset ? Thanks, Souvik
$888 USD 在14天之内
5.0 (2条评论)
4.4
4.4
用户头像
I am a data scientist/quantitative developer with over 5 years of experience doing analysis and prediction. My interest and expertise in using MS Access,R , Matlab ,weka and Database Development.
$555 USD 在10天之内
4.6 (11条评论)
3.0
3.0
用户头像
I have taken a graduate level machine learning course back at UPenn. My preferred platform is Matlab. I am bidding low, because I am new to freelancer, but I believe in myself that I will finish this task in a timely manner.
$250 USD 在2天之内
5.0 (6条评论)
2.7
2.7
用户头像
Hey, I have done projects on matlab regarding machine learning and AI. The best solution to such problems are neural networks or deep learning neural algorithms like the one you have SKYPE translator which can translate languages in real time in context. That's certainly the best algo right now available for deep learning. I can offer you a system with such algorithms. So if you agree we can further discuss in chat....
$500 USD 在10天之内
0.0 (0条评论)
0.0
0.0
用户头像
Hi this is Naveen. I am working as a statistical analyst. I have a master's degree in statistics. I am interested work in such a kind of projects. It would be grateful if you can hire me for this project. Thanking you.
$556 USD 在7天之内
0.0 (0条评论)
0.0
0.0
用户头像
Hi! Could try to generate boolean gate rules using SVM clustering in text miner. From my experience in news parsing, your real task is other, there could not be a statical classifier of sensation. The classes are only regulators of acceleration on bell/S-shaped curve and the classes model is true only for 2 weeks, then need to recalculated. All could be done with your data now - scoring model depending on text class to measure dynamic of possible news propagation speed. But it would be just weak signal detector and there is a big correlation on other events, location and persons (i.e. situation context), so even after detecting signal within model the chances of "news explode" are not strongly predictable. Are source text and topics available in your DB? If not, only propagation model could be built. Could tell you more if you PM me.
$500 USD 在2天之内
0.0 (0条评论)
0.0
0.0
用户头像
Hi, I am very experienced in fesignning and developing pattern classifiers, predictionalmodels, machine learning, etc. I can design the system to perform the tasks you require. I estimate the project can be completed within 5+7 days. I am able to get started immediately and hope to be able to work with you on this project. Sincerely, Michael Cooke, PhD, RIT
$667 USD 在5天之内
0.0 (0条评论)
0.0
0.0
用户头像
A proposal has not yet been provided
$555 USD 在10天之内
0.0 (0条评论)
0.0
0.0
用户头像
This appears to be a reposting of an earlier one on which I had bidded. Before building the classification model, it would be useful to have more information regarding the datasets [ train/test ] and the distribution of the dataset. Estimated Budget : ~ 5000 USD ( Timeline : 25 days ) Price,milestones and timelines flexible and negotiable based on exact project specifications and details. Please refer to my previous bidding in the earlier post for more information.
$5,000 USD 在25天之内
0.0 (0条评论)
0.0
0.0

关于客户

TURKEY的国旗
Turkey
0.0
0
会员自3月 3, 2015起

客户认证

谢谢!我们已通过电子邮件向您发送了索取免费积分的链接。
发送电子邮件时出现问题。请再试一次。
已注册用户 发布工作总数
Freelancer ® is a registered Trademark of Freelancer Technology Pty Limited (ACN 142 189 759)
Copyright © 2024 Freelancer Technology Pty Limited (ACN 142 189 759)
加载预览
授予地理位置权限。
您的登录会话已过期而且您已经登出,请再次登录。