Browsing by Subject "natural language processing"

Now showing 1 - 3 of 3

Defining a Matrix Language in Language Mixing
(2018) Sharath, Vivek
Sentiment analysis on Twitter with stock price and significant keyword correlation
(2013) Zhang, Linhao; Dhillon, Inderjit
Though uninteresting individually, Twitter messages, or tweets, can provide an accurate reflection of public sentiment on when taken in aggregation. In this paper, we primarily examine the effectiveness of various machine learning techniques on providing a positive or negative sentiment on a tweet corpus. Additionally, we apply extracted twitter sentiment to accomplish two tasks. We first look for a correlation between twitter sentiment and stock prices. Secondly, we determine which words in tweets correlate to changes in stock prices by doing a post analysis of price change and tweets. We accomplish this by mining tweets using Twitter's search API and subsequently processing them for analysis. For the task of determining sentiment, we test the effectiveness of three machine learning techniques: Naive Bayes classification, Maximum Entropy classification, and Support Vector Machines. We discover that SVMs give the highest consistent accuracy through cross validation, but not by much. Additionally, we discuss various approaches in training these classifiers. We then apply our findings to on an intra-day market scale to find that there is very little direct correlation between stock prices and tweet sentiment on specifically an intra-day scale. Next, we improve on the keyword search approach by reverse correlating stock prices to individual words in tweets, finding, reasonably, that certain keywords are more correlated with changes in stock prices. Lastly, we discuss various challenges posed by looking at twitter for performing stock predictions.
Update on the Fishes of Texas Project
(2017-03-04) Cohen, Adam; Hendrickson, Dean A.; Urban, Tomislav; Walling, David; Gentle, John; Garrett, Gary; Casarez, Melissa; Martin, F. Douglas
The Fishes of Texas project (www.fishesoftexas.org), originating in 2006, remains the most reliable (quality controlled) and data rich site for acquiring occurrence data for Texas fishes, holding over 124,000 records from 42 institutions. Among many discoveries, the project is responsible for detecting at least 3 freshwater species not previously known from the state. We continue making improvements, but substantial updates so far have been onerous for our developers for various reasons. A recent major update reduces coding redundancies, points the website to a new massively restructured and more fully normalized PostgreSQL database (was MySQL), and places the code in a versioning environment. These changes have little immediate effect on user experience, but will greatly accelerate development. PostgreSQL allows for complex spatial queries which will allow users to quickly map occurrence data alongside many more political/environmental layers than currently possible. While our database/web designers have been implementing these changes and fixing bugs etc., we’ve been preparing resources for them to integrate into the website. Some highlights to expect: 1 new updates to the state Species of Greatest Concern list; 2 expert opinion-determined nativity spatial layers for all freshwater fishes displaying in our new mapping system; 3 dynamic statistical summaries; 4 new data types from the literature (>14,900 records), citizen science (>4,300), anglers (>37,000), and agency databases (>1,000,000); 5 new museum records, many derived from our gap sampling (~19,000, 4 museums); 6 more specimen examinations (>400) and photographs (1000); 7 document archive with “smart” text search tools (currently in beta testing using TPWD fisheries reports). So be patient and keep your eyes open for updates.