Ivan Idris's Blog

March 26, 2016

Social News in 1000 Steps – Step 14

This entry is part 14 of 14 in the series Social NewsI added a table to exclude domains specified by the user. The model and view are as follows: class ExcludeDomain(db.Model): __tablename__ = 'exclude_domains'   id = db.Column(db.Integer, primary_key=True) added … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 26, 2016 13:01

March 25, 2016

Social News in 1000 Steps – Step 13

This entry is part 13 of 13 in the series Social NewsCSV files are great, but of course even a basic database such as SQLite has many added benefits. It’s efficient to start with the most appropriate setup, however even … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 25, 2016 14:26

March 24, 2016

Social News in 1000 Steps – Step 12

This entry is part 12 of 13 in the series Social NewsI setup a prototype front-end, which while not pretty is good enough to do simple CRUD. The setup uses Flask admin (entity classes generated with sqlacodegen). As I said … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 24, 2016 11:36

March 22, 2016

Social News in 1000 Steps – Step 11

This entry is part 11 of 11 in the series Social NewsCode is, of course, very valuable if you are into that sort of thing. High-ranking code n-grams, may be useful for research. To find code I used a regex, … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 22, 2016 13:53

March 21, 2016

Social News in 1000 Steps – Step 10

This entry is part 10 of 10 in the series Social NewsFollowing the previous step, I want to classify the Bing search results. I do this with the TfidfVectorizer and LogisticRegression classes. The results seem to be surprisingly good. The … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 21, 2016 13:42

March 20, 2016

Social News in 1000 Steps – Step 9

This entry is part 9 of 9 in the series Social NewsPirates search with CSE, the Navy searches with Bing (or the other way around). I can compare both in full detail, but you can have your CSE and your … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 20, 2016 11:52

March 18, 2016

Social News in 1000 Steps – Step 8

This entry is part 8 of 8 in the series Social NewsIn SoNaR I want the users to be able to somehow import bookmarks or add bookmarks one at a time. I mock this process with a file. For now … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 18, 2016 11:57

March 16, 2016

Social News in 1000 Steps – Step 7

Apparently Twitter allows you to peform more than 100 queries each 15 minutes. This means that I don’t need to be very selective. Therefore I rewrote about 63 % of the Twitter search code. The new code is as follows: … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 16, 2016 13:32

March 15, 2016

Social News in 1000 Steps – Step 6

This entry is part 6 of 6 in the series Social NewsOnly small improvements this time of which the largest is using Wikipedia to test queries. I query against the live site, however if you are going to query at … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 15, 2016 10:54

March 12, 2016

Social News in 1000 Steps – Step 5

This entry is part 5 of 5 in the series Social NewsTwitter is a (still) popular social media site. We can find relevant information by following informative accounts or by searching Twitter. Today I demonstrate how to search Twitter with … Continue reading →
 •  0 comments  •  flag
Share on Twitter
Published on March 12, 2016 10:31