Editing a network of interconnected concepts

1. A method comprising:

accessing, by one or more computers, a hierarchical network of connected concepts that defines a semantic space, the concepts being connected to one another through connections that have different strength values and different types of relationships, the relationships including (i) a parent-child type of relationship that connects concepts in a first direction in the hierarchical network, (ii) a child-parent type of relationship that connects concepts in a second, different direction in the hierarchical network, and (iii) a lateral bond type of relationship that connects concepts in a third, different direction in the hierarchical network;

identifying multiple connections that define a path between a first concept and a second concept in the hierarchical network;

aggregating the strength values of the multiple connections that define the path;

determining that one or more changes in direction occur along the path;

scaling, by one or more computers, the aggregated strength value using a penalty factor based on determining that one or more changes in direction that occur along the path; and

assigning the scaled strength value as a value that quantifies a proximity of the first concept to the second concept in the semantic space.

2. The method of claim 1 , wherein determining that one or more changes in direction occur along the path comprises determining a quantity of times that the path changes from the first direction or the second direction to the third direction, or vice versa.

3. The method of claim 1 , comprising:

receiving a search query that includes a query term;

determining that the first concept is defined by a plurality of terms including the query term;

determining that the second concept is defined by a plurality of terms including a particular term; and

outputting the particular term as a term that is close in meaning to the query term.

4. The method of claim 1 , wherein aggregating the strength values of the multiple connections that define the path comprises:

aggregating direction-specific strength values of the multiple connections that define the path, in a direction of traversal from the first direction to the second direction.

5. The method of claim 1 , wherein each concept is defined by a synset, and wherein each synset includes a collection of synonyms.

6. The method of claim 1 , comprising:

selecting a scaling factor that increases as the number of connections along the path increases; and

determining the scaled aggregated strength value using the scaling factor.

7. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:

accessing, by one or more computers, a hierarchical network of connected concepts that defines a semantic space, the concepts being connected to one another through connections that have different strength values and different types of relationships, the relationships including (i) a parent-child type of relationship that connects concepts in a first direction in the hierarchical network, (ii) a child-parent type of relationship that connects concepts in a second, different direction in the hierarchical network, and (iii) a lateral bond type of relationship that connects concepts in a third, different direction in the hierarchical network;

identifying multiple connections that define a path between a first concept and a second concept in the hierarchical network;

aggregating the strength values of the multiple connections that define the path;

determining that one or more changes in direction occur along the path;

scaling, by one or more computers, the aggregated strength value using a penalty factor based on determining that one or more changes in direction that occur along the path; and

assigning the scaled strength value as a value that quantifies a proximity of the first concept to the second concept in the semantic space.

8. The system of claim 7 , wherein determining that one or more changes in direction occur along the path comprises determining a quantity of times that the path changes from the first direction or the second direction to the third direction, or vice versa.

9. The system of claim 7 , wherein the operations comprise:

receiving a search query that includes a query term;

determining that the first concept is defined by a plurality of terms including the query term;

determining that the second concept is defined by a plurality of terms including a particular term; and

outputting the particular term as a term that is close in meaning to the query term.

10. The system of claim 7 , wherein aggregating the strength values of the multiple connections that define the path comprises:

aggregating direction-specific strength values of the multiple connections that define the path, in a direction of traversal from the first direction to the second direction.

11. The system of claim 7 , wherein each concept is defined by a synset, and wherein each synset includes a collection of synonyms.

12. The system of claim 7 , wherein the operations comprise:

selecting a scaling factor that increases as the number of connections along the path increases; and

determining the scaled aggregated strength value using the scaling factor.

13. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

accessing, by one or more computers, a hierarchical network of connected concepts that defines a semantic space, the concepts being connected to one another through connections that have different strength values and different types of relationships, the relationships including (i) a parent-child type of relationship that connects concepts in a first direction in the hierarchical network, (ii) a child-parent type of relationship that connects concepts in a second, different direction in the hierarchical network, and (iii) a lateral bond type of relationship that connects concepts in a third, different direction in the hierarchical network;

identifying multiple connections that define a path between a first concept and a second concept in the hierarchical network;

aggregating the strength values of the multiple connections that define the path;

determining that one or more changes in direction occur along the path;

scaling, by one or more computers, the aggregated strength value using a penalty factor based on determining that one or more changes in direction that occur along the path; and

assigning the scaled strength value as a value that quantifies a proximity of the first concept to the second concept in the semantic space.

14. The medium of claim 13 , wherein determining that one or more changes in direction along the path comprises determining a quantity of times that the path changes from the first direction or the second direction to the third direction, or vice versa.

15. The medium of claim 13 , wherein the operations comprise:

receiving a search query that includes a query term;

determining that the first concept is defined by a plurality of terms including the query term;

determining that the second concept is defined by a plurality of terms including a particular term; and

outputting the particular term as a term that is close in meaning to the query term.

16. The medium of claim 13 , wherein aggregating the strength values of the multiple connections that define the path comprises:

aggregating direction-specific strength values of the multiple connections that define the path, in a direction of traversal from the first direction to the second direction.

17. The medium of claim 13 , wherein each concept is defined by a synset, and wherein each synset includes a collection of synonyms.

18. The medium of claim 13 , wherein the operations comprise:

selecting a scaling factor that increases as the number of connections along the path increases; and

determining the scaled aggregated strength value using the scaling factor.