eliminating duplicates in parallel nodes

i have some xml that lists synonyms from several different sources. i'd like to use xsl to output a unique list of these synonyms (ideally each one would also list which resources cite it). suggestions? here's a sample:

<thesaurus1>
<synonym>name A</synonym>
<Alias>name B</Alias>
<Alias>name C</Alias>
</thesaurus1>
<thesaurus2>
<synonym>name D</synonym>
<alias>name E</alias>
<alias>name A</alias>
</thesaurus2>
<thesaurus3>
<synonym>name B</synonym>
<alias>name E</alias>
<alias>name F</alias>
</thesaurus3>
<thesaurus4>
<synonym>name G</synonym>
<alias>name B</alias>
</thesaurus4>

ideally, the output would be an html list like this:

name A (thesaurus1, thesaurus2)
name B (thesaurus1, thesaurus3, thesaurus4)
name C (thesaurus1)
name D (thesaurus2)
name E (thesaurus2, thesaurus3)
name F (thesaurus3)
name G (thesaurus4)

tia,
naomi