Warning: Table './devblogsdb/cache_page' is marked as crashed and last (automatic?) repair failed query: SELECT data, created, headers, expire, serialized FROM cache_page WHERE cid = 'http://www.softdevblogs.com/?q=aggregator/sources/23&page=1' in /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/database.mysql.inc on line 135

Warning: Cannot modify header information - headers already sent by (output started at /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/database.mysql.inc:135) in /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/bootstrap.inc on line 729

Warning: Cannot modify header information - headers already sent by (output started at /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/database.mysql.inc:135) in /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/bootstrap.inc on line 730

Warning: Cannot modify header information - headers already sent by (output started at /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/database.mysql.inc:135) in /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/bootstrap.inc on line 731

Warning: Cannot modify header information - headers already sent by (output started at /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/database.mysql.inc:135) in /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/bootstrap.inc on line 732
Software Development Blogs: Programming, Software Testing, Agile, Project Management
Skip to content

Software Development Blogs: Programming, Software Testing, Agile Project Management

Methods & Tools

Subscribe to Methods & Tools
if you are not afraid to read more than one page to be a smarter software developer, software tester or project manager!

Mark Needham
warning: Cannot modify header information - headers already sent by (output started at /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/database.mysql.inc:135) in /home/content/O/c/n/Ocnarfparking9/html/softdevblogs/includes/common.inc on line 153.
Syndicate content
Thoughts on Software Development
Updated: 5 hours 26 min ago

Neo4j vs Relational: Refactoring – Extracting node/table

Sun, 05/22/2016 - 10:58

In my previous blog post I showed how to add a new property/field to a node with a label/record in a table for a football transfers dataset that I’ve been playing with.

After introducing this ‘nationality’ property I realised that I now had some duplication in the model:

2016 05 22 10 15 15

players.nationality and clubs.country are referring to the same countries but they’ve both got them stored as strings so we can’t ensure the integrity of our countries and ensure that we’re referring to the same country.

We have the same issue in the graph model as well:

2016 05 22 10 40 40

This time Player.nationality and Club.country refer to the same countries.

We can solve our problem by introducing a countries table in the relational model and a set of nodes with a ‘Country’ label in the graph model. Let’s start with relational.

This is the model we’re driving towards:

2016 05 22 10 50 43

The first thing we need to do is create a countries table and populate it:

CREATE TABLE countries (
INSERT INTO countries VALUES('MNE', 'Montenegro');
INSERT INTO countries VALUES('SWZ', 'Swaziland');

Next let’s update the clubs table to reference the countries table:

REFERENCES countries(code);

And let’s run a query to populate that column:

UPDATE clubs AS cl
SET country_id = c.code
FROM clubs
INNER JOIN countries AS c 
ON c.name = clubs.country
WHERE cl.id = clubs.id;

This query iterates over all the clubs, queries the country table to find the country id for that row and then stores it in the ‘country_id’ field. Finally we can remove the ‘country’ field:

DROP COLUMN country;

Now we do the same drill for the players table:

REFERENCES countries(code);
UPDATE players AS p
SET country_id = c.code
FROM players
INNER JOIN countries AS c 
ON c.name = players.nationality
WHERE p.id = players.id;
DROP COLUMN nationality;

Now it’s time for the graph. This is the model we want to get to:

2016 05 22 10 51 49

First we’ll create the countries:

LOAD CSV WITH HEADERS FROM "file:///countries.csv"
AS row
MERGE (country:Country {id: row.countryCode})
ON CREATE SET country.name = row.country

And now let’s get clubs and players to point at those countries nodes and get rid of their respective nationality/country properties:

MATCH (club:Club)
MATCH (country:Country {name: club.country})
MERGE (club)-[:PART_OF]->(country)
REMOVE club.country
MATCH (player:Player)
MATCH (country:Country {name: player.nationality})
MERGE (player)-[:PLAYS_FOR]->(country)
REMOVE player.nationality

And that’s it, we can now write queries against our new model.

Categories: Programming

Neo4j vs Relational: Refactoring – Add a new field/property

Sun, 05/22/2016 - 10:09

A couple of months ago I presented a webinar comparing how you’d model and evolve a data model using a Postgres SQL database and Neo4j.

This is what the two data models looked like after the initial data import and before any refactoring/migration had been done:


2016 05 22 09 49 23


2016 05 22 09 52 16

I wanted to add a ‘nationality’ property to the players table in the SQL schema and to the nodes with the ‘Player’ label in the graph.

This refactoring is quite easy in both models. In the relational database we first run a query to add the ‘nationality’ field to the table:

ALTER TABLE players 
ADD COLUMN nationality VARYING(30);

And then we need to generate UPDATE statements from our data dump to update all the existing records:

UPDATE players 
SET nationality = 'Brazil'
WHERE players.id = '/aldair/profil/spieler/4151';

In the graph we can do this in a single step by processing our data dump using the LOAD CSV command and then setting a property on each player:

LOAD CSV WITH HEADERS FROM "file:///transfers.csv" AS row
MATCH (player:Player {id: row.playerUri})
SET player.nationality = row.playerNationality

If we wanted to make the nationality field non nullable we could go back and run the following queries:

ALTER TABLE players 
ASSERT exists(player.nationality)

And we’re done!

Categories: Programming

R: substr – Getting a vector of positions

Mon, 04/18/2016 - 20:49

I recently found myself writing an R script to extract parts of a string based on a beginning and end index which is reasonably easy using the substr function:

> substr("mark loves graphs", 0, 4)
[1] "mark"

But what if we have a vector of start and end positions?

> substr("mark loves graphs", c(0, 6), c(4, 10))
[1] "mark"

Hmmm that didn’t work as I expected! It turns out we actually need to use the substring function instead which wasn’t initially obvious to me on reading the documentation:

> substring("mark loves graphs", c(0, 6, 12), c(4, 10, 17))
[1] "mark"   "loves"  "graphs"

Easy when you know how!

Categories: Programming

R: tm – Unique words/terms per document

Mon, 04/11/2016 - 06:40

I’ve been doing a bit of text mining over the weekend using the R tm package and I wanted to only count a term once per document which isn’t how it works out the box.

For example let’s say we’re writing a bit of code to calculate the frequency of terms across some documents. We might write the following code:

text = c("I am Mark I am Mark", "Neo4j is cool Neo4j is cool")
corpus = VCorpus(VectorSource(text))
tdm = as.matrix(TermDocumentMatrix(corpus, control = list(wordLengths = c(1, Inf))))
> tdm
Terms   1 2
  am    2 0
  cool  0 2
  i     2 0
  is    0 2
  mark  2 0
  neo4j 0 2
> rowSums(tdm)
   am  cool     i    is  mark neo4j 
    2     2     2     2     2     2

We’ve created a small corpus over a vector which contains two bits of text. On the last line we output a TermDocumentMatrix which shows how frequently each term shows up across the corpus. I had to tweak the default word length of 3 to make sure we could see ‘am’ and ‘cool’.

But we’ve actually got some duplicate terms in each of our documents so we want to get rid of those and only count unique terms per document.

We can achieve that by mapping over the corpus using the tm_map function and then applying a function which returns unique terms. I wrote the following function:

uniqueWords = function(d) {
  return(paste(unique(strsplit(d, " ")[[1]]), collapse = ' '))

We can then apply the function like so:

corpus = tm_map(corpus, content_transformer(uniqueWords))
tdm = as.matrix(TermDocumentMatrix(corpus, control = list(wordLengths = c(1, Inf))))
> tdm
Terms   1 2
  am    1 0
  cool  0 1
  i     1 0
  is    0 1
  mark  1 0
  neo4j 0 1
> rowSums(tdm)
   am  cool     i    is  mark neo4j 
    1     1     1     1     1     1

And now each term is only counted once. Success!

Categories: Programming

Neo4j: A procedure for the SLM clustering algorithm

Sun, 02/28/2016 - 21:40

In the middle of last year I blogged about the Smart Local Moving algorithm which is used for community detection in networks and with the upcoming introduction of procedures in Neo4j I thought it’d be fun to make that code accessible as one.

If you want to grab the code and follow along it’s sitting on the SLM repository on my github.

At the moment the procedure is hard coded to work with a KNOWS relationship between two nodes but that could easily be changed.

To check it’s working correctly I thought it’d make most sense to use the Karate Club data set described on the SLM home page. I think this data set is originally from Networks, Crowds and Markets.

I wrote the following LOAD CSV script to create the graph in Neo4j:

LOAD CSV FROM "file:///Users/markneedham/projects/slm/karate_club_network.txt" as row
MERGE (person1:Person {id: row[0]})
MERGE (person2:Person {id: row[1]})
MERGE (person1)-[:KNOWS]->(person2)

Next we need to call the procedure which will add an appropriate label to each node depending which community it belongs to. This is what the procedure code looks like:

public class ClusterAllTheThings
    public org.neo4j.graphdb.GraphDatabaseService db;
    public Stream<Cluster> knows() throws IOException
        String query = "MATCH (person1:Person)-[r:KNOWS]->(person2:Person) \n" +
                       "RETURN person1.id AS p1, person2.id AS p2, toFloat(1) AS weight";
        Result rows = db.execute( query );
        ModularityOptimizer.ModularityFunction modularityFunction = ModularityOptimizer.ModularityFunction.Standard;
        Network network = Network.create( modularityFunction, rows );
        double resolution = 1.0;
        int nRandomStarts = 1;
        int nIterations = 10;
        long randomSeed = 0;
        double modularity;
        Random random = new Random( randomSeed );
        double resolution2 = modularityFunction.resolution( resolution, network );
        Map<Integer, Node> cluster = new HashMap<>();
        double maxModularity = Double.NEGATIVE_INFINITY;
        for ( int randomStart = 0; randomStart < nRandomStarts; randomStart++ )
            int iteration = 0;
                network.runSmartLocalMovingAlgorithm( resolution2, random );
                modularity = network.calcQualityFunction( resolution2 );
            } while ( (iteration < nIterations) );
            if ( modularity > maxModularity )
                cluster = network.getNodes();
                maxModularity = modularity;
        for ( Map.Entry<Integer, Node> entry : cluster.entrySet() )
            Map<String, Object> params = new HashMap<>();
            params.put("userId", String.valueOf(entry.getKey()));
            db.execute("MATCH (person:Person {id: {userId}})\n" +
                       "SET person:`" + (format( "Community-%d`", entry.getValue().getCluster() )),
        return cluster
                .map( ( entry ) -> new Cluster( entry.getKey(), entry.getValue().getCluster() ) );
    public static class Cluster
        public long id;
        public long clusterId;
        public Cluster( int id, int clusterId )
            this.id = id;
            this.clusterId = clusterId;

I’ve hardcoded some parameters to use defaults which could be exposed through the procedure to allow more control if necessary. The Network#create function assumes it is going to receive a stream of rows containing columns ‘p1’, ‘p2’ and ‘weight’ to represent the ‘source’, ‘destination’ and ‘weight’ of the relationship between them.

We call the procedure like this:

CALL org.neo4j.slm.knows()

It will return each of the nodes and the cluster it’s been assigned to and if we then visualise the network in the neo4j browser we’ll see this:

Graph  1

which is similar to the visualisation from the SLM home page:


If you want to play around with the code feel free. You’ll need to run the following commands to create the JAR for the plugin and deploy it.

$ mvn clean package 
$ cp target/slm-1.0.jar /path/to/neo4j/plugins/ 
$ ./path/to/neo4j/bin/neo4j restart

And you’ll need the latest milestone of Neo4j which has procedures enabled.

Categories: Programming