/*
* Return the most frequent word from the samples whose length is the max word size.
*/
char* most_frequent_word() {
char* word = (char*) malloc((max_word_size+1)*sizeof(char));
Tree_node* max_node = most_frequent_child(prob_root);
Tree_node* node = max_node;
for (int i = 1; i <= max_word_size; i++) {
word[max_word_size-i] = node->symbol;
node = node->parent;
}
return word;
}
/*
* Returns the child of the given node which occurred the most.
* If the depth of the node is max_word_size, then return the node.
*/
Tree_node* most_frequent_child(Tree_node* node) {
if (node_depth(node) == max_word_size) {
return node;
}
int max = 0;
Tree_node* max_node;
Tree_node* current_node = node->child;
while (current_node != NULL) {
Tree_node* max_child = most_frequent_child(current_node);
if (max_child->prob_data != NULL && max_child->prob_data->occurrences > max) {
max = max_child->prob_data->occurrences;
max_node = max_child;
}
current_node = current_node->sibling;
}
return max_node;
}
/*
* Test the result for the bic calculator.
*/
int main(int argc, char** argv) {
char** samples = read_lines(argv[1]);
int depth = strtod(argv[2], NULL);
setup_BIC(samples, depth, prob_root, bic_root);
print_tree(prob_root, -(depth+1));
printf("\n-----------------\n");
printf("Most: %p\n", most_frequent_leaf(prob_root));
printf("most: %p\n", most_frequent_child(prob_root));
printf("\n-----------------\n");
for(Tree_node t = next_node_depth(prob_root, max_word_size); t; t = next_node_depth(t, max_word_size))
printf("%p\n",t);
}